A cross-sectional study investigating the relationship between handgrip strength with multimorbidity in Korean adults: Findings from the Korea National Health and Nutrition Examination Survey (KNHANES) VI-VII (2014–2017)

To date research investigating the associations between handgrip strength and multimorbidity when stratifying by age and sex is limited. Furthermore, this is the rst study in a Korean population, and the rst to include adults ≥ 19 years of age. Here we investigated the associations between handgrip strength and multimorbidity in Korean adults aged ≥ 19 years. In this cross-sectional study data from the Korean National Health and Nutrition Examination Survey were used. Multimorbidity was dened as having two or more chronic diseases. A total of 19,779 participants were eligible for the study (8,730 male and 11,409 female). We grouped and analysed participants based on age and sex (men and women aged 19–49 years, men and women aged 50–80 years) and examined the associations between handgrip strength and multimorbidity using multivariate logistic regressions using stratication by age, sex and body mass index (BMI).


Abstract Background
To date research investigating the associations between handgrip strength and multimorbidity when stratifying by age and sex is limited. Furthermore, this is the rst study in a Korean population, and the rst to include adults ≥ 19 years of age. Here we investigated the associations between handgrip strength and multimorbidity in Korean adults aged ≥ 19 years.

Methods
In this cross-sectional study data from the Korean National Health and Nutrition Examination Survey were used. Multimorbidity was de ned as having two or more chronic diseases. A total of 19,779 participants were eligible for the study (8,730 male and 11,409 female). We grouped and analysed participants based on age and sex (men and women aged 19-49 years, men and women aged 50-80 years) and examined the associations between handgrip strength and multimorbidity using multivariate logistic regressions using strati cation by age, sex and body mass index (BMI).

Results
Multivariate logistic regressions revealed a signi cant inverse association irrespective of sex, between relative handgrip strength and multimorbidity. Participants with the lowest handgrip strength had a signi cantly higher odds ratio (OR) of multimorbidity (men aged 19-49 years: 3.76 and aged 50-80 years: 2.11; women aged 19-49 years: 1.52 and aged 50-80 years: 2.15) when compared to those with highest handgrip strength after adjusting for confounding variables. The highest OR was observed in men aged 19-49 years with a BMI ≥ 25 kg/m 2 .

Conclusion
In the current study we observed a signi cant inverse relationship between handgrip strength and multimorbidity in males and females with two or more chronic diseases. Male participants with a low handgrip strength aged between 19-49 years with a BMI ≥ 25 kg/m 2 may be at greater risk of multimorbidity. This study highlights the need for further longitudinal studies to investigate the effects of increasing handgrip strength combined with weight loss, as an effective strategy to reduce the incidence of multimorbidity Background Multimorbidity can be de ned as the presence of two or more chronic diseases in the same individual that can be clinically managed using medication and other interventions, but not cured [1]. Age has a signi cant effect on the occurrence and incidence of multimorbidity, where 65% of 65-84-year-old's had multimorbidity, when compared with 81.5% of people aged ≥ 85 years in UK [2]. Previous research has demonstrated that patients with multimorbidity are twice as likely to suffer from depression [3], have poor health related quality of life [4] and an increased risk of mortality [5]when compared to those without multimorbidity.
Handgrip strength is commonly used as a measure of physical tness and muscle strength, both of which are related with allcause, cardiovascular and cancer mortality [6]. We provided evidence to substantiate this, where we were observed associations between handgrip strength, type 2 diabetes [7], depression [8] and cardiovascular disease [9]. These data indicate that handgrip strength may be a useful, inexpensive and non-invasive method for assessing multimorbidity.
Poor handgrip strength is commonly associated with a myriad of health problems; however, to date there is a lack of populationbased studies investigating the associations between handgrip strength and multimorbidity. Two cross-sectional studies in Hong-Kong [10] and Brazil [11] both identi ed that men with a low handgrip strength were more likely to have multimorbidity, which was not the case in women. Contrastingly, a cross-sectional study in Germany observed that lower handgrip strength was signi cantly associated with multimorbidity among women but not in men [12]. Interestingly, to date previous research surrounding sex speci c analysis on handgrip strength and multimorbidity have reported inconsistent results. Furthermore, some studies using handgrip strength have failed to correct for body mass index (BMI) and weight and are not a true representation of a national data. Finally, to the best of our knowledge, there are few studies in subjects under the age of 50 years, and the effects of BMI on handgrip strength and multimorbidity remains poorly understood. Therefore, the primary aim of the present study was to examine the association between handgrip strength and multimorbidity in a population-based data generated from the Korean National Health and Nutrition Examination Survey (KNHANES).

Methods
This is a cross-sectional study design of the Korean population from the KNHANES, an annual study that has been conducted since 2007. The study targets only Koreans and is based on a complex strati ed multistage probability cluster survey of a crosssectional and nationally representative sample of individuals from the non-institutionalized civilian population by the Division of Chronic Disease Surveillance, Korea Centres for Disease Control and Prevention.
Data from the KNHANES VI -VII (2014-2017), including data relating to handgrip strength was included. Participants were excluded from the study if they were < 19 years of age (n = 6,386), provided insu cient information of questionnaires relating to multimorbidity (n = 2,975), and had no measurement of handgrip strength for the right or left hand (n = 2,067). Following exclusion of these participants, a total of 19,779 participants (male: 8,730, female: 11,049) remained that were eligible for the study (Fig. 1).

De nitions of multimorbidity
The KNHANES VI-VII questionnaires included 25 chronic diseases (hypertension, dyslipidaemia, stroke, myocardial infarction, angina, osteoarthritis, rheumatic arthritis, tuberculosis, asthma, allergic rhinitis, depression, kidney failure, atopic dermatitis, diabetes mellitus, thyroid disease, stomach cancer, liver cancer, colon cancer, breast cancer, cervical cancer, lung cancer, thyroid cancer, liver cirrhosis, hepatitis B, and hepatitis C). The presence of chronic disease was con rmed by a clinician and con rmed by participants using self-reports. Participants also reported if they still had the disease or if the disease had been treated.
Participants were de ned as multimorbidity if they had two of more con rmed chronic diseases.

Measurement of handgrip strength
Handgrip strength was measured three times in each hand, using a digital grip strength dynamometer (TKK 5401; Takei Scienti c Instruments Co., Ltd., Tokyo, Japan). Trained medical technicians instructed participants to hold the dynamometer with the distal interphalangeal nger joints of the hand at 90° to the handle and to squeeze the handle as rmly as they could whilst being seated. After participants had slowly stood up, handgrip strength was measured during expiration. Study participants conducted three attempts per hand, with a one-minute rest period between each attempt to reduce the effects of fatigue. Handgrip strength of the participants' dominant hand was de ned using questionnaires, and an average of the three measurements was used [13].
Relative handgrip strength was de ned as grip strength per unit of BMI.

Covariates
The self-reported questionnaire provided information relating to participants education, smoking status, alcohol consumption, and physical activity. As education levels are lower in the elderly, education was divided into two categories of ≤ 9 and > 9 years of school. Smoking status was classi ed as never smoking, former smokers and currently smoking. Current smokers self-reported that they smoked > 100 cigarettes in their lifetime and were currently still smoking. Past smokers self-reported that they had a history of smoking. Participants who consumed alcohol were de ned as those who drink at least one glass of alcohol every month over the last year. Physically activity participants were de ned as those who had performed exercise for least 150 minutes per week at a moderate intensity (fast walking, carrying light objects, cleaning, parenting, etc.); 75 minutes per week at a high intensity (lifting or carrying heavy items, digging, labor at construction sites, carrying objects on stairs) (or a combination of medium and high-intensity physical activity (where one minute of high-intensity activity = two minutes of medium intensity activity). BMI was calculated as weight divided by height squared (kg/m 2 ).

Statistical analyses
Statistical guidance was provided by the Korean Centre for Disease Control and Prevention, raw data from the VI-VII (2014)(2015)(2016)(2017) were combined and complex sample analysis was utilised using the weighting data from KNHANES. Sex and age were used to divide participants when investigating associated between handgrip strength and multimorbidity. The groups utilised were  and 50-80 years old, for both male and female participants.
Differences in demographic and anthropometric characteristics were compared using the Student's t-test or Chi-square test, as appropriate. Complex sample multivariate logistic regression was performed to analyse the relationship between relative handgrip strength and multimorbidity, and odds ratio (OR) and 95% con dence intervals (CI) were calculated. Model 1 adjusted for age and Model 2 adjusted for age, education (≤ 9, > 9 years of school), alcohol consumption (yes vs. no), smoking status (never, former, current), and physical activity (yes vs. no). Relative handgrip strength was divided into quartiles. ORs were calculated using complex samples logistic regression analysis to examine associations between the quartile of relative handgrip strength and the presence of multimorbidity based on the highest quartile. Logistic regression analysis was repeated for all age groups using a strati ed approach with a BMI < 25 kg/m 2 and a BMI ≥ 25 kg/m 2 . Statistical signi cance was de ned as P < 0.05. Continuous and categorical variables were expressed as mean ± standard deviation (± SD) and number (n) (%), respectively. All statistical analysis The proportion of females with multimorbidity was higher when compared with males in the 50-80 years olds (female vs. male: 40.17% vs. 31.54%). In contrast, in the 19-49 age group, male participants had a higher proportion of multimorbidity when compared with females (6.21% vs. 5.23%, respectively).
Age, education, smoking status, alcohol consumption, physical activity, number of chronic diseases and dominant handgrip strength were signi cantly different between sexes in the 50-80-year-old group (Table 1). Similar ndings were observed in the 19-49-year-old group; however, age and education were not signi cantly different (Table 1). After adjusting for age, education, physical activity, smoking status, alcohol consumption, relative handgrip strength were inversely associated with multimorbidity in both men and women in both age groups ( Table 2).

BMI = body mass index
Signi cant values were expressed in bold. whereas female participants were similar across all ages (Fig. 2).
In men < 50, the association between handgrip strength and multimorbidity was signi cantly higher in participants with a BMI ≥ 25 kg/m 2 . In men > 50 years old participants with a BMI < and ≥ 25 kg/m 2 were signi cant. (Table 3) The analysis was adjusted for age, education, drinking alcohol, smoking status, and physical activity.

Discussion
The primary aim of the present study was to examine the association between handgrip strength and multimorbidity in a population-based data generated from the Korean National Health and Nutrition Examination Survey (KNHANES). In our study, we observed that the prevalence of multimorbidity increased with age and the proportion of female participants with multimorbidity was higher than male participants aged ≥ 50 years. Similar ndings have previously been reported, where an S-shaped curve was observed for prevalence by age [14]. In the US National Health and Nutrition Examination Survey (NHANES), multimorbidity was higher in female than male participants, and in all adults aged ≥ 20, 60% had two or more multimorbidities compared with 92% in participants aged ≥ 65 years old from 11 possible conditions [15].
The data presented here con rms previously published literature, where handgrip strength and multimorbidity are related when stratifying the analysis by age and sexgender. However, we were not able to nd any effect of sex alone in the general Korean population. The data presented here con rms previously published literature, where handgrip strength and multimorbidity are related when stratifying the analysis by age and gender. However, we were not able to nd any effect of gender alone in the general Korean population. Two previously published cross-sectional studies by Cheung et al. [10] and Amaral et al. [11] in Hong Kong and Brazil, respectively, reported that handgrip strength was an effective marker for identifying male participants with multimorbidity, which may have been due to sex-related hormone differences as no association was observed in females. Contrastingly, Volaklis et al. [12] reported that in a German population handgrip strength was inversely and independently associated with multimorbidity in women ≥ 65 years of age, but not in men. However, in a recent Brazilian study, men without myopenia aged > 60 years with a low handgrip strength were more likely to have multimorbidity [16]. Furthermore, in a recent cross-sectional study conducted across six countries (China, Ghana, India, Mexico, Russia, South Africa), the authors failed to report any sex associations in adults > 50 years of age, but did report an association between low handgrip strength and an increasing number of chronic conditions [17].
Several biological mechanisms have been reported that may be responsible for the relationship between handgrip strength and multimorbidity. For example, the production and release of several cytokines and peptides from skeletal muscle during contractions termed 'myokines' has been associated with metabolic pathways, body fat regulation, anti-in ammatory effects, improved insulin sensitivity, suppression of tumour growth and improved cognitive function [18]. Previous research from our group has also demonstrated a relationship between handgrip strength and the in ammatory marker high sensitivity C-reactive protein [9] Furthermore, reactive oxygen species produced by skeletal muscle are increased during exercise, leading to increased expression of antioxidants [19]. In chronic degenerative diseases such as chronic obstructive pulmonary disease, chronic cardiovascular disease, metabolic diseases, and neurodegeneration have all been shown to be affected by chronic oxidative stress. [20], suggesting a potential protective effect of exercise The ndings presented here demonstrate that males in the 19-49 years old group with a BMI ≥ 25 kg/m 2 combined with a low handgrip strength had the highest OR. Supporting these ndings, in a Canadian study population, the incidence of multimorbidity was higher in obese participants, but not in those with normal BMI [21]. Obesity and a lack of physical activity have previously been identi ed to be a signi cant predictor of multimorbidity in adults aged > 55 years in New Zealand [22]. Furthermore, in a cohort study from Iran, obese participants had an OR that was 2.33-fold higher than participants with a normal BMI [23]. Obesity is a well-established risk factor for chronic diseases, and has been shown to cause systemic in ammation and activate well-known pathways for cancer, cardiovascular disease and type 2 diabetes mellitus [24]. The results from the current study amongst others may provide the basis for further management of participants with a high BMI and low handgrip strength for the early onset of non-communicable diseases and multimorbidity.
The present study has several important strengths. Firstly, this is the rst study to investigate the relationship between handgrip strength and multimorbidity in a large nationally representative sample size from Korea. Secondly, we were able to stratify the analysis between handgrip strength and multimorbidity by age and gender. Thirdly, this is the rst study to investigate the impact of handgrip strength on multimorbidity in adults between 19 and 80 years of age. Finally, we were able to identify that a combination of an increased BMI and low handgrip strength had a more pronounced effect on multimorbidity in 19-49-year-olds.
This study is not without limitations. Firstly, all participants were Korean, meaning that our ndings cannot be extrapolated to other ethnicities. Secondly, information relating to medical conditions was based on self-reporting questionnaires, which may cause recall biases. Finally, this study is cross-sectional and any attempt to determine causality or temporality in the association between handgrip and multimorbidity is not possible.

Conclusion
In the current study we observed a signi cant inverse relationship between handgrip strength and multimorbidity in males and females with two or more chronic diseases. Male participants with a low handgrip strength aged between 19-49 years with a BMI ≥ 25 kg/m 2 may be at greater risk of multimorbidity. This study highlights the need for further longitudinal studies to investigate the effects of increasing handgrip strength combined with weight loss, as an effective strategy to reduce the incidence of multimorbidity.
Abbreviations BMI: body mass index, CI = con dence interval, KNHANES = Korean National Health and Nutrition Examination Survey, OR = odds ratio, SD = standard deviation Declarations Figure 1 Participant flow diagram illustrating the number of patients excluded by various criteria, and the number of patient data sets analysed.

Figure 2
Multivariate logistic regression association between relative handgrip strength quartile and risk of multimorbidity. The analysis was adjusted for age, education, alcohol consumption, smoking status, and physical activity.