1 Low handgrip strength is closely associated with anemia ...42 1.35-3.34) and for those aged ≥65 years (OR=1.92, 95% CI: 1.42-2.58). 43 Conclusion 44 This study showed a strong

1

1 Low handgrip strength is closely associated with anemia among adults: A cross-

2 sectional study using Korea National Health and Nutrition Examination Survey

3 (KNHANES)

4 Short title: Low handgrip strength is closely associated with anemia among adults

5 Yu-mi Gi1, Boyoung Jung 2*, Koh-Woon Kim3, Jae-Heung Cho4, In-Hyuk Ha 2*

6

7 1 Jaseng Hospital of Korean Medicine, 536 Gangnam-daero, Gangnam-gu, Seoul 06110, Republic

8 of Korea

9 2Jaseng Spine and Joint Research Institute, Jaseng Medical Foundation, 3F, 538 Gangnam-daero,

10 Gangnam-gu, Seoul 06110, Republic of Korea

11 3Department of Korean Rehabilitation Medicine, College of Korean Medicine, Kyung Hee

12 University, 26 Kyungheedae-ro, Dongdaemun-gu, Seoul 130-701, Republic of Korea

13 4Department of Korean Rehabilitation Medicine, College of Korean Medicine, Kyung Hee

14 University, 26 Kyungheedae-ro, Dongdaemun-gu, Seoul 130-701, Republic of Korea

15 *Corresponding authors

16 E-mail: [email protected] (BJ)

17 E-mail: [email protected] (IHH)

18

.CC-BY 4.0 International licensenot certified by peer review) is the author/funder. It is made available under aThe copyright holder for this preprint (which wasthis version posted May 29, 2019. . https://doi.org/10.1101/652545doi: bioRxiv preprint

https://doi.org/10.1101/652545

http://creativecommons.org/licenses/by/4.0/

2

19 ABSTRACT

20 Background21 Anemia occurs because of insufficient hemoglobin, which provides oxygen to the body. Because

22 of its close relationship with various illnesses, it must always be investigated clinically. Previous

23 studies have demonstrated an association between hemoglobin concentration and handgrip

24 strength. Thus, we aimed to analyze the association between handgrip strength and anemia in

25 Korean adults to determine whether the handgrip strength test can be used as a tool to identify

26 anemia.

27 Methods28 The research subjects’ data were extracted from the 6th and 7th Korean National Health and

29 Nutrition Examination Survey, between January 2013 and December 2017. Overall, data of 16,637

30 adults (weighted n= 9,734,598) were analyzed. Differences in sociodemographic factors (sex, age,

31 education, income, and employment), lifestyle factors (alcohol consumption, smoking, and

32 physical activity), and illness and health factors (body mass index, vitamin intake, iron intake,

33 comorbid illnesses, and handgrip strength) by existence of anemia were analyzed using a chi

34 square test. Binary logistic regression was performed to identify factors of anemia. Subgroup

35 analysis, stratified by sex and age, was performed.


https://doi.org/10.1101/652545


3

36 Results37 Among Korean adults aged ≥19 years, 745,296 (7.7%) had anemia. Higher odds ratio (OR) of

38 anemia occurred in the weak handgrip strength group compared to the strong handgrip strength

39 group (OR=1.92, 95% CI: 1.58-2.33). The subgroup analysis showed a higher OR for anemia in the

40 weak handgrip strength group than in the strong handgrip strength group, regardless of sex or

41 age. However, the results showed that this association was greater for males (OR=2.13, 95% CI:

42 1.35-3.34) and for those aged ≥65 years (OR=1.92, 95% CI: 1.42-2.58).

43 Conclusion44 This study showed a strong association between handgrip strength and anemia, which was

45 particularly strong for males and those aged ≥65 years. Therefore, it is anticipated that handgrip

46 strength can be used in anemia screening tests as a useful tool.

47


https://doi.org/10.1101/652545


4

48 Introduction 49 Anemia occurs because of insufficient hemoglobin, which provides oxygen to the body. It is one

50 of the major clinical problems affecting about 12% of the elderly [1]. The World Health

51 Organization (WHO) defines anemia as hemoglobin values below 12 g/dl for females and 13 g/dl

52 for males [2]. Anemia affects mortality rate [3], disability [4], the decline in physical activity level [4,

53 5], and decline in the quality of life [6] in the elderly. Furthermore, since anemia has a close

54 relationship with many illnesses, it is recommended that when hemoglobin level is lower than the

55 normal, it should always be followed-up clinically [5]. Hence, the prevalence of anemia can then

56 be reduced by appropriate treatment or by preventing the underlying illnesses that are related to

57 aging [7].

58 Muscle weakness is related to poor physical activity and incident mobility limitations in the

59 elderly [3, 6, 8-11]. Handgrip strength, a simple, cost-effective tool used to measure strength, is

60 utilized in the diagnosis of muscle weakness in epidemiological studies [10]. Handgrip strength is

61 also a non-invasive and cheap tool that can help in identifying those with a higher likelihood of

62 anemia, and can be used as a tool to identify anemia in the community [12]. Some research has

63 reported that handgrip strength in middle-age (45-65 years) is related to functional limitations

64 and disabilities at least 25 years later [3, 11, 13]. Therefore, handgrip strength has been

65 considered a factor predicting healthy aging [11]. Handgrip strength is also related to the


https://doi.org/10.1101/652545


5

66 strength of other muscle groups, and therefore can be a good index to represent overall strength

67 [14]. Therefore, low handgrip strength signifies low muscle mass and poor clinical mobility [15].

68 Oxidative stress is one of the most important mechanism to explain the occurrence of age-

69 related illnesses [16] including the decline in strength [17]. Because hemoglobin concentration

70 plays an important role in the oxidation of blood [18, 19], anemia, defined by hemoglobin,

71 represents a decline in the oxygenation function and can therefore be related to low strength [4].

72 Aging is the most frequent cause of reduced muscle strength [20], and anemia is commonly

73 related to aging [21]. Previous research in Australian elderly found a direct relationship between

74 reduction in hemoglobin level and the reduction in handgrip strength [22]. In addition, in a

75 research among elderly participants aged ≥65 years, anemia patients, defined by the hemoglobin

76 level, had significantly lower handgrip strength compared to patients without anemia [4]. From

77 these previous studies, it is evident that the hemoglobin level can be directly related to handgrip

78 strength and that this is especially pronounced in the elderly population; however, the mechanism

79 underlying this is unclear. Moreover, research on the association between handgrip strength and

80 anemia in Korean adults has provided insufficient data for a clear conclusion. Therefore, in the

81 present study, we aimed to determine whether an association exists between handgrip strength

82 and anemia based on the data from the 6th and 7th Korean National Health and Nutrition

83 Examination Survey (KNHANES).


https://doi.org/10.1101/652545


6

84 Materials and Methods

85 Study design and population86 The present study used the data from the 6th and 7th KNHANES that was conducted from

87 January 2013 to December 2017 [17]. KNHANES is a 3-yearly cross-sectional research, which

88 includes examinations, health surveys, and nutritional surveys, and involves a representative

89 sample of the entire Korean population, using a stratified cluster-sampling method.

90 The KNHANES study sample consisted of Korea citizens (residents in group facilities, such as

91 nursing homes, military, and prison as well as foreigners are excluded). To increase the

92 representativeness of the sample and the accuracy of assumptions, sampling was designed to

93 distinguish between province and city, neighborhood, towns, and townships, and the types of

94 home, to extract sampling regions (investigated district). We used the data from 2013-2017 to

95 measure handgrip strength. The subject exclusion criteria were as follows: missing data on anemia

96 (n=8,928) and on handgrip strength (n=7,556), those aged under 19 years (n=2,307), and other

97 missing data (n=3,797). Ultimately, the analyses were performed using data of 16,637 subjects.

98 Final analysis set considering weight 9,734,598 subjects (Fig 1).

99 Fig 1. Subjects’ flow diagram.

100


https://doi.org/10.1101/652545


7

101 Outcomes and other variables

102 Measurement of handgrip strength

103 Handgrip strength, an indicator of muscle strength, was evaluated using a handgrip strength test

104 according to the recommendations of the Institute of Medicine [14]. For the standard value of

105 handgrip strength, we used the standard for the skeletal muscle mass (26 kg for men, 16 kg for

106 women) suggested by the North American Foundation for the National Institutes of Health

107 Sarcopenia Project published in 2015 [22]. The handgrip strength for each hand was measured

108 three times using a digital grip strength dynamometer (TKK 5401; Takei Scientific Instruments Co,

109 Ltd, Tokyo, Japan). Trained medical technicians instructed the study subjects while in a sitting

110 position to hold the dynamometer with the distal interphalangeal finger joints at a 90° angle to

111 the handle, and to squeeze the handle as firmly as they could. Another handgrip strength

112 measurement was done during expiration, after the subject had slowly stood up. Study subjects

113 conducted 3 attempts per hand, with a 1-minute rest period between each attempt to reduce the

114 effect of fatigue due to repetition. The value for the handgrip strength measurements was the

115 average of the three measurements, for either hand [23]. Using this mean handgrip strength

116 values, subjects were categorized into the weak (<26 kg for men, <16 kg for women) and strong

117 (≥26 kg for men, ≥16 kg for women) groups.

118


https://doi.org/10.1101/652545


8

119 Measurement of hemoglobin (Hb) levels

120 Hemoglobin (Hb) was used as an index to determine anemia. In accordance with the WHO

121 standards, male and female subjects with Hb less than 13 and 12 g/dL, respectively, were defined

122 as anemia patients [13].

123 Description of demographic and other characteristics of the study

124 population

125 We analyzed the general characteristics, socioeconomic background characteristics, and lifestyle

126 habits of the patients. Body mass index (BMI), a value derived by dividing the weight with the

127 square of height, was categorized into <18.5 kg/m2 (underweight), 18.5-24.9 kg/m2 (normal), and

128 ≥25.0 kg/m2 (overweight) [16]. For smoking, subjects were categorized into nonsmoker, ex-

129 smoker, or current smoker. The frequency of alcohol consumption was divided into four

130 categories (‘none,’ ‘less than 1 time per month’, ‘1-4 times per month’, and ‘more than 5 times

131 per month.’) For employment, subjects were categorized into ‘unemployed (student, housewife,

132 etc.)’ or ‘employed.’ For marital status, subjects were categorized into three groups ('married-

133 cohabiting' for those who had spouses or cohabiting partners; 'married-non cohabiting, bereaved,

134 or divorced’ for those who were living apart, bereaved, divorced, etc.; or 'unmarried' for those

135 who were unmarried). Household income was categorized into four groups (low, low-moderate,

136 moderate-high, and high). Educational level was also categorized into four groups (elementary


https://doi.org/10.1101/652545


9

137 school or less, middle school, high school, and college or over). Level of daily workout refers to

138 the frequency of strength exercise per week, and was categorized into four groups ('none'; 'light'

139 [1 day]; 'moderate' [2~3 days]; and 'heavy/extreme' [4 days or more]).

140 Description of nutritional characteristics and comorbidity in the

141 study population

142 For nutritional characteristics, iron intake, vitamin A intake, and vitamin C intake per day were

143 categorized as ‘insufficient’ if the intake was lower than the recommended level by sex and age

144 and ‘sufficient’ if the intake was above the recommended level. All blood samples were drawn

145 after fasting and were analyzed within 24 hours of being drawn. The 2015 standards of the

146 Dietary Reference Intakes for Koreans (KDRIs) were used for iron and vitamin C intake per day. For

147 vitamin A intake, the 2010 KDRIs standards [24] were used with identical units (μg RE/day) for

148 KHANES and KDRIs [25] (μg RAE/day), in 2015. Vitamin A intake per day for males and females

149 aged 19-49 and ≥50 years were 750 and 700 μg RE/day; and 650 and 600 μg RE/day, respectively.

150 Vitamin C intake per day was 100 mg for both males and females over 19 years old. Iron intake

151 per day for males were 10 and 9 mg/day for those aged 19-64 and ≥65 years, respectively; while

152 for females aged 19-49, 50-74, and ≥75 years, these were 14, 8, and 7 mg/day, respectively.

153 For comorbid diseases, hypertension, cancer, hyperlipidemia, cardiovascular disease, kidney

154 disease, bone disease, and thyroid disease were included. Subjects were categorized into a ‘yes’ or


https://doi.org/10.1101/652545


10

155 ‘no’ group based on the diagnosis or absence of at least one of these diseases by a doctor.

156 Statistical analysis157 Because KNHANES data were obtained from a complex sampling design, all statistical analyses

158 methods used complex sample analyses considering weight, stratified variables, and cluster

159 sampling. Differences in sociodemographic factors (sex, age, education, income, and employment),

160 lifestyle factors (alcohol consumption, smoking, and physical activity), and illness and health

161 factors (BMI, vitamin intake, iron intake, comorbid illnesses, and handgrip strength) due to

162 presence of anemia were analyzed using a complex sample analysis of chi square (χ2) test.

163 Categorical variables were presented as proportions (n, %) using a χ2 test of independence, while

164 continuous variables were presented with estimated value ± standard error (SE) using the general

165 linear model.

166 Chi-square test was performed on risk factors associated with anemia. Furthermore, to identify

167 factors associated with anemia, odds ratio (OR) and 95% confidence intervals (95% CI) were

168 obtained through a binary logistic regression analysis after setting anemia as the dependent

169 variable and handgrip strength as the main independent variable, while controlling for

170 sociodemographic factors, lifestyle habit factors, and disease-related and health factors. The

171 association between handgrip strength and anemia were subgroup-analyzed by sex (male/female)

172 and age (younger or older than 65). All statistical analyses were performed using SPSS version


https://doi.org/10.1101/652545


11

173 25.0 (SPSS Inc., Chicago, IL, USA) and SAS version 9.4 (SAS Institute Inc, Cary, NC), and the

174 significance level of all tests were defined as p-value less than 0.05.

175 Ethics statement176 The KNHANES 6th and 7th were conducted by the Korea Center for Disease Control and

177 Prevention (KCDC). All survey protocols were approved by the institutional review board (IRB) of

178 the KCDC (approval numbers: 2013-07CON-03-4C, 2013-12EXP-03-5C, and 2015-01-02-6C).

179 Informed consent was obtained from all participants when the surveys were conducted. Original

180 data are publicly available for free in the KNHANES website (http://knhanes.cdc.go.kr) for

181 purposes such as academic research. Approval of IRB was not required because the study did not

182 deal with any sensitive information, but rather accessed only publicly available data from the

183 KNHANES (JASENG IRB File No. 2019-03-009).

184

185 Result186 Table 1 shows the general characteristics of the subjects. Patients with anemia were 7.7%

187 overall. Majority of subjects (51.2%) were female, of which, 11.9% had anemia. In males only 3.2%

188 had anemia. Overall, 6.7% of subjects and 15.8% were in the strong and weak handgrip strength

189 groups, respectively (p<0.0001). The mean age of subjects with anemia was higher (50.27 years)


https://doi.org/10.1101/652545


12

190 than for those in the group without anemia (46.14 years) (p<0.0001). By marital status, those in

191 the married-not cohabiting, bereaved, or divorced group had a higher percentage (11.6%) of

192 anemia than those in the married-cohabiting (8.1%) and the unmarried (4.7%) groups (p<0.0001).

193 The prevalence rates of anemia in the unemployed and employed groups were 10% and 6.3%,

194 respectively (p<0.0001).

195 By individual lifestyle habit factors, by the level of daily workout, in the light (1 day) daily

196 workout group 615,680 (8.5%) subjects had anemia, and had higher prevalence of anemia than

197 those in the moderate (5.7%) and heavy/extreme (4.9%) daily workout groups, (p<0.0001). The

198 prevalence rate of anemia in the underweight group (11.4%) was higher than that in the normal

199 weight or overweight groups (8.6% and 5.5%, respectively) (p<0.0001).

200 For the nutritional and disease-related factors, the prevalence rate of anemia in the group with

201 comorbid diseases was 2.5% higher than in the group without (p<0.0001). For iron intake, the

202 prevalence rate of anemia was 3.5% greater in the insufficient group compared to the sufficient

203 group (p<0.0001). For vitamin A intake, the prevalence rates of anemia was also higher (8.1%) for

204 the insufficient group than for the sufficient group (6.5%) (p=0.001). However, no difference

205 between groups was found in vitamin C intake per day (p=0.505).

206 Table 1 Characteristics of the study population

207


https://doi.org/10.1101/652545


13

AnemiaP-

value†

No YesCharacteristics (%)

Number (n)

Percent (%)

Number (n)

Percent (%)

Total (100.0) 8,989,302 92.3 745,296 7.7Handgrip strength

(kg) Weak 10.1 825,860 84.2 154,758 15.8

Strong 89.9 8,164,451 93.3 590,537 6.70.000

Gender Male 48.8 4,593,800 96.8 152,732 3.2(Weighted %) Female 51.2 4,396,511 88.1 592,563 11.9 0.000

Age (Mean±SE) 46.14 (±0.230) 50.27 (±0.580) 0.000

Marital status Married-cohabiting 66.7 10,759 91.9 1,038 8.1Married-not cohabiting, bereaved, or

divorced

10.1 1,910 88.4 267 11.6

Unmarried 23.2 2,525 95.3 139 4.7

0.000

Household Low 24.6 3,617 92.2 356 7.8income Lower middle 24.7 3,745 90.9 412 9.1

Higher middle 25.1 3,875 92.7 355 7.3(Weighted %) High 25.6 3,957 93.5 321 6.5

0.001

Unemployed 36.3 3,178,617 90.0 352,480 10.0

EmploymentEmployed 63.7 5,811,694 93.7 392,816 6.3 0.000

Educational level, n (%)

Elementary school or less

15.0 1,297,532 89.1 158,564 10.9

Middle school 8.6 780,173 93.4 54,719 6.6

High school 35.8 3,226,791 92.6 258,853 7.4

College or over 40.6 3,681,699 93.1 271,921 6.9

0.000

Alcohol Never drink 23.0 1,976,814 88.4 259,504 11.6

consumption less than 1 time per month 18.1 1,601,112 90.6 165,298 9.4

(weighted %) 1-4 times per month 35.7 3,246,613 93.4 229,017 6.6≥5 drinking

episodes per month 23.2 2,165,772 95.9 91,476 4.1

0.000

Smoking Non-smoker 58.1 5,080,850 89.8 575,101 10.2

Past smoker 21.2 1,947,273 94.3 117,502 5.7

Current smoker 20.7 1,962,187 97.4 52,692 2.6

0.000

Level of daily Light 74.4 6,629,928 91.5 615,680 8.5 0.000


https://doi.org/10.1101/652545


14

workout Moderate 9.2 840,670 94.3 51,197 5.7

Heavy/extreme 16.4 1,519,713 95.1 78,419 4.9

BMI (kg/m2) Normal weight 62.1 5,526,034 91.4 517,041 8.6mean (SD) Underweight 4.3 367,291 88.6 47,138 11.4

Overweight 33.7 3,096,986 94.5 181,116 5.50.000

Comorbid diseases, No 65.8 5,974,940 93.2 435,246 6.8

n (%) Yes 34.2 3,015,371 90.7 310,049 9.30.000

Iron intake per Insufficient 37.2 3,243,939 89.5 379,999 10.5

day(mg/day) Sufficient 62.8 5,746,372 94.0 365,296 6.0 0.000

Vitamin A intake Insufficient 72.6 6,492,478 91.9 572,688 8.1per day

(μg RE/day) Sufficient 27.4 2,497,833 93.5 172,607 6.50.001

Vitamin C intake per day (mg/day) Insufficient 71.0 6,379,979 92.3 535,787 7.7

Sufficient 29.0 2,610,332 92.6 209,508 7.40.505

208 * Other diseases include hypertension, cancer, hyperlipidemia, cardiovascular disease, kidney disease, bone 209 disease, and thyroid disease

210 † Chi-Square test or t-test was performed to determine differences between groups with/without anemia.

211 Abbreviations: SD, standard deviation; BMI, body mass index;

212

213 Table 2 shows the characteristic of subjects by sex. For handgrip strength, in both males (17.0%

214 vs. 3.1%) and females (16.5% vs. 10.7%) the prevalence rate was significantly higher in the weak

215 group than in the strong group (p<0.0001). Only in males were the mean ages of persons with

216 and without anemia with a difference of about 18 years. Similarly, anemia prevalence rate was

217 only higher in the underweight male group (15.6%) compared to the normal (5.8%) and

218 overweight (3.1%) groups (p<0.0001), unlike for females (p=0.074). Anemia prevalence rate was

219 also higher only in males with comorbid diseases (9.4%) than for those without comorbid disease

220 (2.0%) (p<0.0001); but not for females, although the group without comorbid disease (11.8%) had


https://doi.org/10.1101/652545


15

221 slightly higher prevalence rate than those with comorbid diseases (10.9%) (p=0.065). The

222 prevalence rate of anemia was lower in the iron intake sufficient groups (4.8% for males and 9.7%

223 for females) compared to the insufficient groups (5.7% for males and 13.6% for females), but this

224 was not significant in males (p=0.492).


https://doi.org/10.1101/652545


16

Table 2. Characteristics of the study population by anemia according to sex

Male Female

anemia anemia

No Yes No YesTotal(n)Number

(n)Percent

(%)Number

(n)Percent

(%)

P-value† Total(n)

Number (n)

Percent (%)

Number (n)

Percent (%)

P-value†

Total 4746532 4593800 96.8 152732 3.2 4989075 4396511 88.1 592563 11.9

weak 436455 378554 86.7 57901 13.3 0.000 544164 447307 82.2 96858 17.8 0.000Hand grip(kg)-2Q

strong 4310077 4215246 97.8 94831 2.2 4444910 3949205 88.8 495706 11.2

45.25±0.278 63.62±0.914 0.000 47.07±0.260 46.82±0.609 0.000Age(yr) (Mean±SE)45.84±0.279 47.04±0.252

Urban area 3971297 3848194 96.9 123103 3.1 0.192 4263914 3762553 88.2 501361 11.8 0.503RegionRural area 775235 745606 96.2 29629 3.8 725161 633959 87.4 91202 12.6

Married-cohabiting 3160785 3038119 96.1 122666 3.9 0.000 3330775 2929202 87.9 401573 12.1 0.146Married-not cohabiting, bereaved, or divorced 240072 222432 92.7 17640 7.3 745488 648825 87.0 96663 13.0Marital status

Unmarried 1345675 1333249 99.1 12426 0.9 912811 818485 89.7 94327 10.3Low 1180706 1135023 96.1 45682 3.9 0.115 1216663 1075659 88.4 141005 11.6 0.003

Lower middle 1160131 1120798 96.6 39334 3.4 1242594 1064273 85.6 178321 14.4

Higher middle 1190316 1153349 96.9 36967 3.1 1255929 1114039 88.7 141889 11.3Income

High 1215379 1184630 97.5 30749 2.5 1273889 1142541 89.7 131349 10.3

Unemployed 1162602 1091803 93.9 70799 6.1 0.000 2368495 2086815 88.1 281680 11.9 0.972Employment

Employed 3583930 3501997 97.7 81933 2.3 2620580 2309697 88.1 310883 11.9Elementary school or

less 504239 454972 90.2 49267 9.8 0.000 951858 842561 88.5 109297 11.5 0.014Middle school 403036 384138 95.3 18898 4.7 431856 396034 91.7 35822 8.3High school 1770921 1718534 97.0 52387 3.0 1714723 1508257 88.0 206465 12.0

Education Level

College or over 2066023 2034615 98.5 31408 1.5 1887597 1647084 87.3 240513 12.7


https://doi.org/10.1101/652545


17

Never drink 700037 644773 92.1 55264 7.9 0.000 1536281 1332041 86.7 204240 13.3 0.011less than 1 time per

month 563460 545632 96.8 17828 3.2 1202949 1055480 87.7 147469 12.3

1-4 times per month 1824263 1781506 97.7 42758 2.3 1651367 1465107 88.7 186260 11.3

Alcohol consumption

≥5 times per month 1658772 1621890 97.8 36882 2.2 598477 543883 90.9 54594 9.1Non-smoker 1244812 1208848 97.1 35964 2.9 0.000 4411140 3872002 87.8 539137 12.2 0.029Past smoker 1766677 1680389 95.1 86287 4.9 298099 266884 89.5 31215 10.5Smoking

Current smoker 1735043 1704562 98.2 30481 1.8 279836 257625 92.1 22211 7.9Light 3142034 3024438 96.3 117595 3.7 0.000 4103574 3605490 87.9 498084 12.1 0.143

Moderate 531443 524745 98.7 6697 1.3 360424 315925 87.7 44499 12.3Level of daily

workoutHeavy/extreme 1073055 1044616 97.3 28439 2.7 525076 475097 90.5 49980 9.5

High 1205557 1183127 98.1 22430 1.9 0.000 1459232 1283937 88.0 175295 12.0 0.982

Middle 2777411 2688397 96.8 89013 3.2 2801742 2470974 88.2 330768 11.8

Low 746842 706690 94.6 40152 5.4 705094 620951 88.1 84143 11.9Stress level

Zero 16722 15585 93.2 1136 6.8 23007 20649 89.8 2357 10.2Normal weight 2725568 2623997 96.3 101571 3.7 0.000 3317508 2902038 87.5 415470 12.5 0.074Underweight 125363 112066 89.4 13297 10.6 289066 255225 88.3 33842 11.7BMI (kg/m2)Overweight 1895601 1857737 98.0 37864 2.0 1382501 1239249 89.6 143252 10.4

No 3242557 3200355 98.7 42201 1.3 0.000 3167629 2774584 87.6 393045 12.4 0.065other diseases

Yes 1503975 1393444 92.7 110531 7.3 1821445 1621927 89.0 199519 11.0

insufficient 1117883 1079199 96.5 38684 3.5 0.492 2506055 2164739 86.4 341315 13.6 0.000Iron intake per day(mg/day) sufficient 3628649 3514600 96.9 114048 3.1 2483020 2231772 89.9 251248 10.1

insufficient 3362075 3243391 96.5 118685 3.5 0.029 3703091 3249087 87.7 454004 12.3 0.087Vitamin A intake per day (μg RE/day) sufficient 1384457 1350409 97.5 34047 2.5 1285984 1147424 89.2 138560 10.8

insufficient 3444984 3338160 96.9 106824 3.1 0.357 3470782 3041819 87.6 428963 12.4 0.048Vitamin C intake per day (mg/day) sufficient 1301548 1255640 96.5 45908 3.5 1518293 1354692 89.2 163601 10.8


https://doi.org/10.1101/652545


18

* Other diseases include hypertension, cancer, hyperlipidemia, cardiovascular disease, kidney disease, bone disease, and thyroid disease† Chi-Square test or t-test was performed to determine differences between groups with/without anemia according to sex.

Abbreviations: yr, year; SE, standard error; BMI, body mass index;


https://doi.org/10.1101/652545


19

226 Table 3 shows the characteristics of subjects by age and anemia status. The mean age of

227 subjects aged <65 and ≥65 years were 41.72 and 72.41 years, respectively (p<0.0001). In the <65

228 and ≥65 years, respectively, anemia prevalence rate was higher in the weak than the strong

229 handgrip strength group (<65 years, 10.9 vs. 6.5%, p<0.0001) and (≥65 years, 20.2% vs. 8.7%,

230 p<0.0001), respectively. For the <65-year-olds, more female than males had anemia (11.6% vs.

231 1.7%) and a slight difference for the ≥65-year-olds (13.1% vs. 12.0%), respectively. Significant (<65

232 years, 8.7% vs. 5.9%, p<0.0001) and non-significant (≥65 years, p=0.207) difference occurred in

233 the prevalence rate of anemia in the unemployed and employed groups, respectively. For alcohol

234 consumption, the prevalence rate of anemia was lower for those who consume alcohol and

235 especially those with a higher frequency of alcohol consumption, regardless of age (p<0.0001).

236 For the level of daily workout, the prevalence rate of anemia was higher in the light group

237 compared to the moderate or heavy/extreme group for all subjects regardless of age (p<0.0001).

238 Also for BMI, the prevalence rate of anemia in the underweight group (<65 years, 10.1%; ≥65

239 years, 23.2%) was far higher than the prevalence rate among the normal weight or overweight

240 groups (p<0.0001), regardless of age. Furthermore, for iron intake, the prevalence rate of anemia

241 in the insufficient group (<65 years 9.7%; ≥65 years 15.7%), was higher than the prevalence rate

242 in the sufficient group (<65 years 4.9%; ≥65 years, 11.3%) for both age groups.


https://doi.org/10.1101/652545


20

Table 3 Characteristics of the study population by anemia according to ageage <65 65≤age

anemia anemiaNo Yes No YesTotal(n)

Number (n)

Percent (%)

Number (n)

Percent (%)

P-value† Total(n)

Number (n)

Percent (%)

Number (n)

Percent (%)

P-value†

Total 8233180 7677209 93.2 555971 6.8 1501418 1312093 87.4 189325 12.6weak 467750 416722 89.1 51028 10.9 0.000 512869 409138 79.8 103730 20.2 0.000Hand grip(kg)-2Q strong 7766439 7261496 93.5 504943 6.5 988549 902955 91.3 85594 8.7Male 4060656 3990031 98.3 70625 1.7 0.000 685876 603768 88.0 82107 12.0 0.309

GenderFemale 4173533 3688187 88.4 485346 11.6 815542 708325 86.9 107217 13.1

41.68±0.177 42.20±0.460 0.000 72.18±0.101 73.96±0.258 0.000Age(yr) (Mean±SE) 41.72±0.171 72.41±0.096Urban area 7090203 6608349 93.2 481853 6.8 0.700 1145008 1002397 87.5 142611 12.5 0.671RegionRural area 1143986 1069869 93.5 74118 6.5 356410 309696 86.9 46714 13.1Married-

cohabiting 5485389 5077323 92.6 408066 7.4 0.000 1006171 889997 88.5 116174 11.5 0.019

Bereaved or divorced 500145 458235 91.6 41911 8.4 485415 413022 85.1 72393 14.9Marital status

Unmarried 2248655 2142660 95.3 105995 4.7 9831 9074 92.3 758 7.7Low 2033962 1891883 93.0 142078 7.0 0.012 363407 318799 87.7 44609 12.3 0.042

Lower middle 2039062 1878252 92.1 160810 7.9 363663 306819 84.4 56844 15.6Higher middle 2080731 1944130 93.4 136601 6.6 365513 323259 88.4 42255 11.6Income

High 2080434 1963953 94.4 116481 5.6 408834 363217 88.8 45617 11.2Unemployed 2516091 2296543 91.3 219547 8.7 0.000 1015006 882074 86.9 132932 13.1 0.207

EmploymentEmployed 5718098 5381675 94.1 336424 5.9 486412 430019 88.4 56392 11.6Elementary

school or less 583431 547142 93.8 36288 6.2 0.471 872666 750390 86.0 122276 14.0 0.005

Middle school 617782 583334 94.4 34447 5.6 217111 196839 90.7 20272 9.3High school 3238009 3010854 93.0 227155 7.0 247635 215937 87.2 31698 12.8

Education Level

College or over 3791628 3534320 93.2 257308 6.8 161992 147379 91.0 14613 9.0


https://doi.org/10.1101/652545


21

Never drink 1527997 1378194 90.2 149803 9.8 0.000 708321 598620 84.5 109701 15.5 0.000less than 1 time

per month 1545539 1407307 91.1 138233 8.9 220870 193805 87.7 27065 12.3

1-4 times per month 3181318 2982231 93.7 199087 6.3 294312 264382 89.8 29930 10.2

Alcohol consumption

≥5 times per month 1979335 1910487 96.5 68848 3.5 277914 255286 91.9 22628 8.1

Non-smoker 4760771 4297891 90.3 462880 9.7 0.000 895181 782959 87.5 112222 12.5 0.093Past smoker 1622582 1566527 96.5 56055 3.5 442194 380746 86.1 61448 13.9Smoking

Current smoker 1850836 1813800 98.0 37037 2.0 164043 148388 90.5 15655 9.5Light 6042466 5591838 92.5 450629 7.5 0.000 1203141 1038090 86.3 165051 13.7 0.000

Moderate 830371 782488 94.2 47883 5.8 61496 58183 94.6 3313 5.4Level of daily workout

Heavy/extreme 1361352 1303893 95.8 57459 4.2 236780 215820 91.1 20960 8.9High 2404389 2243643 93.3 160746 6.7 0.554 260400 223421 85.8 36979 14.2 0.534

Middle 4838237 4506540 93.1 331696 6.9 740916 652831 88.1 88085 11.9Low 973175 909957 93.5 63218 6.5 478761 417684 87.2 61077 12.8

Stress level

Zero 18388 18077 98.3 311 1.7 21340 18157 85.1 3183 14.9Normal weight 5135795 4747771 92.4 388024 7.6 0.000 907281 778263 85.8 129018 14.2 0.000Underweight 375286 337220 89.9 38066 10.1 39143 30071 76.8 9072 23.2BMI (kg/m2)Overweight 2723108 2593227 95.2 129881 4.8 554994 503759 90.8 51235 9.2

No 6092026 5689324 93.4 402703 6.6 0.317 318160 285616 89.8 32544 10.2 0.040other diseases

Yes 2142163 1988894 92.8 153268 7.2 1183258 1026477 86.8 156781 13.2insufficient 3168779 2860360 90.3 308418 9.7 0.000 455159 383578 84.3 71581 15.7 0.001Iron intake per

day(mg/day) sufficient 5065410 4817858 95.1 247553 4.9 1046258 928515 88.7 117744 11.3insufficient 5891215 5468415 92.8 422800 7.2 0.004 1173951 1024062 87.2 149889 12.8 0.639Vitamin A intake

per day (μg RE/day) sufficient 2342974 2209803 94.3 133171 5.7 327467 288031 88.0 39436 12.0

insufficient 5869086 5475024 93.3 394063 6.7 0.799 1046680 904956 86.5 141724 13.5 0.025Vitamin C intake per day (mg/day) sufficient 2365102 2203194 93.2 161908 6.8 454738 407137 89.5 47600 10.5

* Other diseases include hypertension, cancer, hyperlipidemia, cardiovascular disease, kidney disease, bone disease, and thyroid disease† Chi-Square test or t-test was performed to determine differences between groups with/without anemia according to age (age <65/ 65≤age).


https://doi.org/10.1101/652545


22

Abbreviations: yr, year; SE, standard error; BMI, body mass index;


https://doi.org/10.1101/652545


23

244 Table 4 shows the association between handgrip strength and anemia using logistic regression

245 analyses. Model 1 adjusted for sociodemographic factors of anemia, such as sex, age, region, and

246 income. Model 2 adjusted for individual lifestyle habit factors, such as smoking, alcohol

247 consumption, and BMI, in addition to the sociodemographic factors. Finally, Model 3 shows

248 association with anemia after adjusting for sociodemographic factors, individual lifestyle habit

249 factors, and nutritional and disease-related factors.

250 The ORs of anemia for weak handgrip strength compared to the strong handgrip strength was

251 2.02, 1.98, and 1.92 in Models 1, 2, and 3, respectively, showing a significant association between

252 weak handgrip strength and anemia (p<0.0001). The odds of having anemia were least for

253 persons with 1-4 times per month frequency of alcohol consumption compared to the ≥5

254 drinking episodes per month persons, and highest in the light level of daily workout compared to

255 the heavy/extreme groups. The odds ratio of underweight and overweight BMI was higher

256 compared to the normal weight, and the odds of anemia was also higher in persons with

257 comorbid diseases compared to persons without comorbid diseases at 1.12. Also, the prevalence

258 of anemia was higher for persons with insufficient iron, vitamin A, and vitamin C intake per day

259 compared to persons with sufficient intakes (Table 4).

260 Table 5 demonstrates the association between handgrip strength and anemia stratified by sex

261 and age. As in Table 4, Model 1 shows the association with anemia while controlling for


https://doi.org/10.1101/652545


24

262 sociodemographic factors; Model 2 shows the values after controlling for individual lifestyle habit

263 factors in addition to factors in Model 1; and Model 3 shows the values after controlling for

264 sociodemographic factors, individual lifestyle habit factors, and nutritional and disease-related

265 factors. Regardless of sex and age, there was a significant association between handgrip strength

266 and anemia. In particular, the trend of weak handgrip strength in those with anemia compared to

267 those without anemia was more pronounced in males than females (p=0.001, 95% CI: 1.35-3.34);

268 and in those aged ≥65 years compared to those <65 (p<0.0001, 95% CI: 1.42-2.58).

269


https://doi.org/10.1101/652545


25

Table 4. Association between hand grip and anemia

Total Model 1 Model 2 Model 3Category

9,734,598 OR 95% CI P-value OR 95% CI P-

value OR 95% CI P-value

weak 980,619 2.02 1.67 2.45 0.000 1.98 1.63 2.40 0.000 1.92 1.58 2.33 0.000Hand grip (kg)-2Qstrong 8,754,988 1.00 1.00 1.00Male 4,746,532 1.00 1.00Gender

Female 4,989,075 2.45 1.94 3.10 0.000 2.22 1.71 2.88 0.000 2.08 1.60 2.70 0.000Age (years) 1.01 1.00 1.02 0.003 1.01 1.00 1.02 0.029 1.01 1.00 1.02 0.021

Urban area 8,235,210 1.00 0.82 1.23 0.978 1.01 0.82 1.24 0.954 1.01 0.82 1.24 0.939RegionRural area 1,500,396 1.00 1.00

Married-cohabit 6,491,560 1.33 1.01 1.74 0.041 1.37 1.04 1.80 0.027 1.40 1.06 1.84 0.019Married-no co habit or bereaved

or divorced 985,560 1.23 0.88 1.70 0.221 1.33 0.95 1.85 0.093 1.33 0.95 1.86 0.091Marital status

Unmarried 2,258,486 1.00 1.00 1.00Low 2,397,369 1.27 1.04 1.56 0.020 1.29 1.05 1.59 0.014 1.26 1.02 1.55 0.030

Lower middle 2,402,725 1.52 1.27 1.83 0.000 1.52 1.27 1.83 0.000 1.50 1.24 1.80 0.000Higher middle 2,446,245 1.18 0.97 1.43 0.090 1.19 0.98 1.44 0.083 1.16 0.96 1.42 0.125

Income

High 2,489,268 1.00 1.00 1.00Unemployed 3,531,097 1.06 0.91 1.22 0.461 0.97 0.84 1.12 0.668 0.98 0.85 1.13 0.764

Employment Employed 6,204,510 1.00 1.00 1.00Elementary school or less 1,456,097 0.69 0.55 0.86 0.001 0.74 0.59 0.94 0.013 0.71 0.56 0.91 0.006

Middle school 834,892 0.60 0.46 0.79 0.000 0.64 0.49 0.84 0.001 0.64 0.49 0.85 0.002High school 3,485,644 0.98 0.83 1.15 0.796 1.02 0.87 1.20 0.791 1.02 0.87 1.20 0.831

Education Level

College or over 3,953,620 1.00 1.00 1.00Never drink 2,236,318 1.49 1.21 1.84 0.000 1.52 1.23 1.87 0.000

less than 1 time per month 1,766,409 1.39 1.10 1.76 0.006 1.42 1.12 1.80 0.0031-4 times per month 3,475,630 1.26 1.03 1.56 0.026 1.28 1.04 1.58 0.019

Alcohol consumption

≥5 drinking episodes per month 2,257,249 1.00 1.00


https://doi.org/10.1101/652545


26

Non-smoker 5,655,951 1.60 1.20 2.13 0.001 1.62 1.22 2.16 0.001Past smoker 2,064,776 1.95 1.46 2.60 0.000 1.94 1.45 2.59 0.000Smoking

Current smoker 2,014,880 1.00 1.00

High 7,245,607 0.78 0.33 1.82 0.557 0.76 0.32 1.81 0.537Middle 891,867 0.80 0.34 1.87 0.601 0.79 0.33 1.87 0.586

Low 0.82 0.35 1.93 0.645 0.81 0.34 1.94 0.634Stress level

Zero 1,598,132 1.00 1.00Light 1.26 1.01 1.57 0.044 1.25 1.00 1.56 0.054

Moderate 1.14 0.82 1.58 0.433 1.15 0.83 1.60 0.392Level of daily workoutHeavy/extreme 1.00 1.00Normal weight 6,043,076 1.00 1.00Underweight 414,429 1.40 1.20 1.63 0.000 1.41 1.21 1.65 0.000BMI (kg/m2)Overweight 3,278,102 1.74 1.25 2.44 0.001 1.77 1.26 2.48 0.001

No 6,410,186 1.00Comorbid diseases *Yes 3,325,421 1.12 0.95 1.31 0.176

insufficient 3,623,938 1.42 1.22 1.65 0.000Iron intake per day(mg/day)sufficient 6,111,669 1.00

insufficient 7,065,166 1.02 0.87 1.21 0.785Vitamin A intake per day (μg RE/day) sufficient 2,670,440 1.00

insufficient 6,915,766 1.03 0.90 1.19 0.632Vitamin C intake per day (mg/day) sufficient 2,819,840 1.00

Notes: Logistic regression analysis with complex sampling design was performed by adjusting for covariates. OR indicates odds ratio; 95% CI, 95% confidence interval.Model 1 was adjusted for sex, age, region, marital status, income, employment, educationModel 2 was adjusted for Model 1 + alcohol consumption, smoking, level of daily workout, stress level and BMI Model 3 was adjusted for Model 2 + other disease, iron intake per day, vitamin A intake per day, and vitamin C intake per day

270


https://doi.org/10.1101/652545


27

Table 5. Association between hand grip and anemia by sub group

Model 1 Model 2 Model 3Class Category

OR 95% CI P-value OR 95% CI P-value OR 95% CI P-valueweak 2.36 1.53 3.63 0.000 2.11 1.34 3.31 0.001 2.13 1.35 3.34 0.001

Male Hand grip (kg)-2Qstrong 1.00 1.00 1.00

weak 1.79 1.41 2.27 0.000 1.77 1.40 2.25 0.000 1.71 1.35 2.16 0.000Female Hand grip (kg)-2Q

strong 1.00 1.00 1.00

weak 1.59 1.18 2.15 0.003 1.58 1.17 2.14 0.003 1.57 1.16 2.12 0.003age <65 Hand grip (kg)-2Q

strong 1.00 1.00 1.00

weak 1.97 1.47 2.64 0.000 1.90 1.41 2.55 0.000 1.92 1.42 2.58 0.00065≤ age Hand grip (kg)-2Q

strong 1.00 1.00 1.00

Notes: Logistic regression analysis with a complex sampling design was performed by adjusting for covariates. OR indicate odds ratio; 95% CI, 95% confidence interval.Model 1 was adjusted for sex, age, region, marital status, income, employment, educationModel 2 was adjusted for Model 1 + alcohol consumption, smoking, level of daily workout, stress level and BMI Model 3 was adjusted for Model 2 + other disease, iron intake per day, vitamin A intake per day, and vitamin C intake per day

271

272


https://doi.org/10.1101/652545


28

273 Discussion274 This study analyzed the association between handgrip strength and anemia in a total of

275 9,734,598 adults >19 years using a representative, reliable data that represents all Korean citizens.

276 As a result, we found that handgrip strength and anemia had a clear correlation in the Korean

277 adult population >19 years old. When handgrip strength was categorized into weak or strong

278 based on the standards of the National Institutes of Health, the proportion of those having

279 anemia with weaker hand grip strength was higher. Even after controlling for sociodemographic

280 factors such as sex, age, region, and income, there was a clear correlation between handgrip

281 strength and anemia. Furthermore, after controlling for individual lifestyle habit factors, such as

282 smoking, alcohol consumption, nutritional and disease-related factors, handgrip strength and

283 anemia showed a very significant association (OR=1.92, 95% CI: 1.58-2.33). Between the group

284 with anemia and without anemia, indices such as BMI, level of daily workout, and iron intake per

285 day showed significant differences. A peculiar finding is that the prevalence rate of anemia was

286 higher for non-smokers and those with lower consumption of alcohol (Table 1). This is deduced

287 to be due to the fact that females form a greater proportion of the anemia patients at 79.5%

288 compared to males.

289 The relationship between anemia and handgrip strength can be explained by a biological

290 mechanism as reported in previous studies. The low strength of anemia patients can be explained


https://doi.org/10.1101/652545


29

291 to be due to weakness and fatigue, and this can increase the risk of disability in these individuals

292 [26]. Hemoglobin, in the red blood cells, is responsible for transporting oxygen to various parts of

293 the body. Therefore, the lower blood concentration of hemoglobin means lower oxygen is

294 transported, and this can increase the risk of various mobility disorders along with fatigue [27].

295 Because the hemoglobin concentration plays an important role in the oxygenation of the blood

296 [18, 19], anemia defined by the level of hemoglobin represents reduction of oxygenation function.

297 Hypoxia due to anemia can reduce the oxygen supply to the muscles and thereby affect the

298 strength of the muscles [4].

299 We performed subgroup analyses by sex and age, which were predicted to affect anemia. In the

300 subgroup by sex, the correlation between handgrip strength and anemia was higher for males

301 compared to females. For males, indices such as BMI, level of daily workout, and comorbid

302 disease had significant relationships with anemia (Table 2). Additionally, in the subgroups by age,

303 the association between handgrip strength and anemia was more pronounced in subjects ≥65

304 years compared to those <65 years Although the proportion of females among anemia patients

305 was far higher in the group <65 years, the male or female proportions did not show large

306 differences in those ≥65 years. It was also found that BMI, level of daily workout, and iron intake

307 per day were significantly related to anemia in both groups (Table 3).

308 Previous studies have shown significant relationships between anemia and handgrip strength in


https://doi.org/10.1101/652545


30

309 only females, rather than males. These include a study which found significant differences in

310 hemoglobin levels in males among the South African population >40 years old [28] and a study

311 which found an association between anemia and handgrip strength in Brazilian female population

312 >60 years old [12]. However, this study showed higher odds of anemia in the weak handgrip

313 strength group compared to the strong handgrip strength group in both males and females, and

314 this association was greater in the male compared to the female group. The reason that the

315 differences in the prevalence rates of anemia of males and females were different in each study

316 can be the heterogeneity of the aging population with regards to race, life environment, and

317 health problems, and these can all affect the hemoglobin levels [29].

318 Furthermore, previous studies on the elderly have found a clear association between handgrip

319 strength and anemia. According to the research by Hirani V et al., the decline in hemoglobin

320 concentration in the Australian elderly was directly related to the decline in handgrip strength

321 [30]). Pennix et al also reported that anemia patients (defined by the hemoglobin level among

322 elderly subjects >65 years), had far lower handgrip strength compared to patients without anemia

323 [4]. Moreover, research on the association between anemia and handgrip strength in the elderly

324 population >90 years old [28, 31] and research that found an association between anemia,

325 handgrip strength, and lower-body strength in the elderly >100 years old [32] showed that the

326 association between handgrip strength and association is large in the elderly population,


https://doi.org/10.1101/652545


31

327 especially as age increases. The reason could be that the reduction in handgrip strength due to

328 aging can represent the overall health of the elderly [33]. In particular, because the elderly often

329 have higher blood concentration of C-reactive protein, acute or chronic inflammation can result in

330 an even larger physical decline [4, 5]. Therefore, the maintenance and management of handgrip

331 strength are important especially in the elderly population, and it will be important to identify

332 factors that affect handgrip strength [34].

333

334 There are several limitations with this research. First, because this study used a cross-sectional

335 study design based on the KNHANES, caution must be taken to interpret a causal relationship

336 between the variables. In other words, we cannot interpret as a causal relationship that handgrip

337 strength is an independent causative factor of anemia or an epiphenomenon factor. Secondly,

338 there is a limitation in identifying the risk factors of anemia because of insufficient nutritional data

339 on folic acid or vitamin B12, which can cause anemia, since the effect of these nutritional vitamins

340 on anemia could not be considered. It was difficult to clearly consider the effect of diseases that

341 could be related to anemia. Thirdly, because KNHANES only considered Hb levels to define

342 anemia, it was difficult to consider the types or causes of anemia. Therefore, additional research is

343 necessary on various variables that can affect anemia; it is anticipated that the results of such a

344 study can be used for clinical purposes. Despite these limitations, this research has the following


https://doi.org/10.1101/652545


32

345 advantages. First, this study is the first research to examine the correlation between handgrip

346 strength and anemia in Korean adults. Although the fact that the concentration of hemoglobin

347 can be directly related to handgrip strength has been found through previous studies, there was

348 still insufficient research to confirm the association between anemia and handgrip strength.

349 Because this research used data on the general population of Korea, the reliability of the data is

350 high. Secondly, the results of this study showed that the relationship between handgrip strength

351 and anemia was very strong; thus, it was suggested that handgrip strength can be used as a

352 useful measurement tool for screening for anemia. Through this study, it will be possible to check

353 and clinically follow up on the population with high likelihood of anemia through handgrip

354 strength measurement, which is a simple tool that represents strength.

355 Conclusion356 The results of the present study showed that there was a strong association between handgrip

357 strength and anemia. In particular, the association between handgrip strength and anemia was

358 more pronounced for males and those who are older than 65. Therefore, handgrip strength can

359 be used as a risk factor when screening for anemia in patients, and can be used in clinical

360 practice for testing for anemia in the elderly, in epidemiological studies, or during treatments.


https://doi.org/10.1101/652545


33

361 Acknowledgements362 Disclosure of interests

363 The authors have declared that no competing interests exist.

364 Author Contributions

365 Conceptualization, Resources: IHH BJ. Data curation: BJ. Formal analysis, Visualization, Writing

366 (Original draft): YMG BJ. Methodology, Project administration, validation: KWK JHC. Software: BJ

367 Supervision: IHH. Investigation, Writing (Review&Editing): YMG BYJ IHH

368

369 Reference370 1. Salive ME, Cornoni‐Huntley J, Guralnik JM, Phillips CL, Wallace RB, Ostfeld AM, et al.

371 Anemia and hemoglobin levels in older persons: relationship with age, gender, and health status.

372 Journal of the American Geriatrics Society. 1992;40(5):489-96.

373 2. Organization WH. Nutritional anemia: report of a WHO Scientific Group. Geneva: World

374 Health Organization. 1968.

375 3. Ferrucci L, Guralnik JM, Buchner D, Kasper J, Lamb SE, Simonsick EM, et al. Departures

376 from linearity in the relationship between measures of muscular strength and physical

377 performance of the lower extremities: the Women's Health and Aging Study. The Journals of

378 Gerontology Series A: Biological Sciences and Medical Sciences. 1997;52(5):M275-M85.

379 4. Penninx BW, Pahor M, Cesari M, Corsi AM, Woodman RC, Bandinelli S, et al. Anemia is

380 associated with disability and decreased physical performance and muscle strength in the elderly.

381 Journal of the American Geriatrics Society. 2004;52(5):719-24.

382 5. Penninx BW, Guralnik JM, Onder G, Ferrucci L, Wallace RB, Pahor M. Anemia and decline

383 in physical performance among older persons. The American journal of medicine. 2003;115(2):104-

384 10.


https://doi.org/10.1101/652545


34

385 6. Sallinen J, Stenholm S, Rantanen T, Heliövaara M, Sainio P, Koskinen S. Hand‐grip

386 strength cut points to screen older persons at risk for mobility limitation. Journal of the American

387 Geriatrics Society. 2010;58(9):1721-6.

388 7. Gualandro SF, Hojaij N, Jacob Filho W. Deficiência de ferro no idoso. Rev Bras Hematol

389 Hemoter. 2010;32(supl 2):57-61.

390 8. Manini TM, Clark BC. Dynapenia and aging: an update. Journals of Gerontology Series A:

391 Biomedical Sciences and Medical Sciences. 2011;67(1):28-40.

392 9. Manini TM, Visser M, Won‐Park S, Patel KV, Strotmeyer ES, Chen H, et al. Knee extension

393 strength cutpoints for maintaining mobility. Journal of the American Geriatrics Society.

394 2007;55(3):451-7.

395 10. Hicks GE, Shardell M, Alley DE, Miller RR, Bandinelli S, Guralnik J, et al. Absolute strength

396 and loss of strength as predictors of mobility decline in older adults: the InCHIANTI study.

397 Journals of Gerontology: Series A: Biomedical Sciences and Medical Sciences. 2011;67(1):66-73.

398 11. Lauretani F, Russo CR, Bandinelli S, Bartali B, Cavazzini C, Di Iorio A, et al. Age-associated

399 changes in skeletal muscles and their effect on mobility: an operational diagnosis of sarcopenia.

400 Journal of applied physiology. 2003;95(5):1851-60.

401 12. Santos PHS, Carmo ÉA, Carneiro JAO, Nery AA, Casotti CA. Handgrip strength: An

402 effective screening instrument for anemia in the elderly women. Public Health Nursing.

403 2019;36(2):178-83.

404 13. Chung H, Moon H, Song B, Kim M. Hemoglobin, hematocrit and serum ferritin as

405 markers of iron status. Korean J Nutr. 1991;24(5):450-7.

406 14. Pate RR, Daniels S. Institute of Medicine report on fitness measures and health outcomes

407 in youth. JAMA pediatrics. 2013;167(3):221-2.

408 15. Yoo J-I, Ha Y-C, Kwon H-B, Lee Y-K, Koo K-H, Yoo M-J. High prevalence of sarcopenia in

409 Korean patients after hip fracture: a case-control study. Journal of Korean medical science.

410 2016;31(9):1479-84.

411 16. Razak F, Anand SS, Shannon H, Vuksan V, Davis B, Jacobs R, et al. Defining obesity cut

412 points in a multiethnic population. Circulation. 2007;115(16):2111.

413 17. Ahn S. Korean national health and nutrition examination survey: MS Thesis; 2005.

414 18. Giulivi C, Davies K. A novel antioxidant role for hemoglobin. The comproportionation of

415 ferrylhemoglobin with oxyhemoglobin. Journal of Biological Chemistry. 1990;265(32):19453-60.

416 19. Grune T. Oxidative stress, aging and the proteasomal system. Biogerontology.

417 2000;1(1):31-40.

418 20. Irace C, Scarinci F, Scorcia V, Bruzzichessi D, Fiorentino R, Randazzo G, et al. Association

419 among low whole blood viscosity, haematocrit, haemoglobin and diabetic retinopathy in subjects

420 with type 2 diabetes. British Journal of Ophthalmology. 2011;95(1):94-8.

421 21. Kim J-H. The Presence of Diabetes Mellitus and Anemia in Korean Adults-based on data


https://doi.org/10.1101/652545


35

422 from 2005 Korean National Health and Nutrition Examination Survey (KNHANES III). Journal of

423 Nutrition and Health. 2008;41(6):502-9.

424 22. Alley DE, Shardell MD, Peters KW, McLean RR, Dam T-TL, Kenny AM, et al. Grip strength

425 cutpoints for the identification of clinically relevant weakness. Journals of Gerontology Series A:

426 Biomedical Sciences and Medical Sciences. 2014;69(5):559-66.

427 23. Roberts HC, Denison HJ, Martin HJ, Patel HP, Syddall H, Cooper C, et al. A review of the

428 measurement of grip strength in clinical and epidemiological studies: towards a standardised

429 approach. Age and ageing. 2011;40(4):423-9.

430 24. Society MoHaWTKN. Dietary Reference Intakes for Koreans 2010. 2010.

431 25. Society MoHaWTKN. Dietary Reference Intakes for Koreans 2015. Seoul: The Korean

432 Nutrition Society. 2015.

433 26. Fukushima T, Nakano J, Ishii S, Natsuzako A, Kawachi H, Sakamoto J, et al. Influence of

434 Hemoglobin Level on Muscle and Physical Functions, Activities of Daily Living, and Quality of Life

435 in Patients With Hematological Malignancies. Integrative Cancer Therapies.

436 2019;18:1534735419842196.

437 27. Cho S-B. Effect of Muscular Reinforcement Exercise of Serum Albumin, WBC and

438 Hemoglobin in the Elderly Women The journals of the Korean Society Of Sports Science.

439 2013;22(4):1075-81.

440 28. Payne CF, Davies JI, Gomez-Olive FX, Hands KJ, Kahn K, Kobayashi LC, et al. Cross-

441 sectional relationship between haemoglobin concentration and measures of physical and cognitive

442 function in an older rural South African population. J Epidemiol Community Health.

443 2018;72(9):796-802.

444 29. Eisenstaedt R, Penninx BW, Woodman RC. Anemia in the elderly: current understanding

445 and emerging concepts. Blood reviews. 2006;20(4):213-26.

446 30. Hirani V, Naganathan V, Blyth F, Le Couteur DG, Seibel MJ, Waite LM, et al. Low

447 hemoglobin concentrations are associated with sarcopenia, physical performance, and disability in

448 older Australian men in cross-sectional and longitudinal analysis: the Concord Health and Ageing

449 in Men project. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences.

450 2016;71(12):1667-75.

451 31. Thein M, Ershler WB, Artz AS, Tecson J, Robinson BE, Rothstein G, et al. Diminished

452 quality of life and physical function in community-dwelling elderly with anemia. Medicine.

453 2009;88(2):107.

454 32. Haslam A, Hausman DB, Davey A, Cress ME, Johnson MA, Poon LW. Associations of

455 anemia and physical function in Georgia centenarians. Journal of the American Geriatrics Society.

456 2012;60(12):2362.

457 33. Rijk JM, Roos PR, Deckx L, van den Akker M, Buntinx F. Prognostic value of handgrip

458 strength in people aged 60 years and older: A systematic review and meta‐analysis. Geriatrics &


https://doi.org/10.1101/652545


36

459 gerontology international. 2016;16(1):5-20.

460 34. Lee J, Hong YS, Park S-H, Kang KY. High serum uric acid level is associated with greater

461 handgrip strength in the aged population. Arthritis research & therapy. 2019;21(1):73.

462


https://doi.org/10.1101/652545



https://doi.org/10.1101/652545


Documents

1 Low handgrip strength is closely associated with anemia ...42 1.35-3.34) and for those aged ≥65 years (OR=1.92, 95% CI: 1.42-2.58). 43 Conclusion 44 This study showed a strong