According to recent data from the World Health Organization, approximately 800 million adults worldwide currently have diabetes, which is about four times the number reported in 1990 [1]. In Korea, 16.7% of adults aged 30 years or older were diagnosed with diabetes during 2021-2022. Among them, 65.8% were aware of their condition and were receiving treatment, but only 9.7% reached the integrated management goals for blood glucose, blood pressure, and lipids at the same time [2]. These results show that diabetes is not only an individual health problem but also a social and economic problem, and they suggest that preventive management strategies need to be strengthened.

Early diagnosis is one of the main ways to reduce the burden of diabetes, because it allows effective management of the disease [3]. Lifestyle modification is also a core strategy for delaying the onset and progression of diabetes. Regular exercise and physical activity (PA) help with blood glucose control, insulin sensitivity, and weight management, and they reduce the risk of complications [4-6]. For early detection to be effective in real settings, indicators that can identify high-risk individuals are needed, and PA level and physical fitness indicators can be useful indicators for preventive management. Among fitness indicators, handgrip strength (HGS) and fat-free mass (FFM) are considered key indicators because they are closely associated with prevalence of diabetes [7,8]. Studies have consistently reported an inverse association between HGS and the incidence and prevalence of diabetes and metabolic disease [7,9].

HGS is generally classified as absolute handgrip strength (AHGS) or relative handgrip strength (RHGS). AHGS reflects muscular function itself, but it is greatly influenced by physique, body weight, and sex [10,11]. RHGS is calculated by dividing AHGS by body weight or body mass index (BMI). Because RHGS adjusts for differences in physique, it is suggested as an indicator that better reflects metabolic vulnerability [12]. Based on these characteristics, RHGS may be a more valid indicator for screening and assessing the risk of diabetes and metabolic syndrome.

FFM is defined as a body composition indicator that represents the total non-adipose tissue of the body [13]. FFM is used as a surrogate indicator for muscle mass. Higher FFM or muscle mass has been associated with a lower risk of incident type 2 diabetes, and with better metabolic control and prognosis in patients with established diabetes [8,14]. Recent meta-analysis showed that the people with diabetes had a 2.38 times higher probability of having lower muscle mass compared with the non-diabetes group [15]. In the Korean Genome and Epidemiology Study, participants in the lowest tertile of weight-adjusted appendicular skeletal muscle mass index (appendicular skeletal muscle mass/body weight) had a 2.05 times higher risk of incident type 2 diabetes compared with those in the highest tertile (follow-up: 9.06 years; multivariable-adjusted HR 2.05, 95% CI 1.73-2.43). The inverse association was maintained even after additional adjustment for body fat, waist circumference, and BMI [16]. Among Japanese Americans with low BMI, larger thigh circumference was significantly associated with a lower 10-year risk of incident diabetes [17]. Based on these findings, lower-body muscle mass, which represents the largest skeletal muscle group in the human body, may be theoretically more closely related to insulin resistance and glucose homeostasis.

Most previous studies have examined only a single muscle-related indicator, such as HGS or total FFM. As a result, the independent role of lower-body FFM as a regional muscle indicator remains unclear. In addition, many studies did not adjust enough for important confounders. Examples include body composition (body fat percentage, waist circumference, BMI), lifestyle factors (PA, diet, smoking, alcohol drinking), and socioeconomic factors (income, education level). When the adjustment is insufficient, the effect estimates can be biased. Furthermore, few studies have simultaneously examined absolute and relative handgrip strength, as well as total and lower-body FFM, within the same study population and statistical framework, or directly compared their relative explanatory power. This study aimed to examine the independent and comparative associations of AHGS, RHGS, total FFM, and lower-body FFM with diabetes mellitus prevalence using nationally representative data from the 2022-2023 Korea National Health and Nutrition Examination Survey (KNHANES).

The present study compared the associations of HGS indicators (absolute and relative) and FFM indicators (total and lower-body) with the odds of diabetes using KNHANES 2022-2023 data. In the total sample, the odds of diabetes were 61% lower in the highest RHGS tertile and 43% lower in the highest lower-body FFM tertile compared with the lowest tertiles in Model 2. In sex-stratified analyses, an inverse association between RHGS and the odds of diabetes was observed in both men and women.

In the present study, even within the same sex, RHGS showed a more consistent and clear inverse association with the odds of diabetes than AHGS. AHGS reflects muscular function directly, but it is greatly influenced by sex and physique, including height and weight [11]. In contrast, RHGS, calculated as AHGS divided by body weight or BMI, adjusts for differences in physique and may more clearly capture the metabolic signal of muscular function itself. In line with this, large cohort studies and meta-analyses have reported that AHGS is inversely associated with incident type 2 diabetes, and weight-standardized RHGS has been suggested as a valid metabolic risk marker [9]. According to a study using UK Biobank data, participants in the lowest RHGS tertile had a higher risk of incident diabetes than those in the highest RHGS tertile, with hazard ratios of 2.22 (95% CI: 1.84-2.67) in men and 1.96 (95% CI: 1.52-2.53) in women. For AHGS, participants in the lowest tertile also had a higher risk of incident diabetes than those in the highest tertile, but the magnitude of association was smaller, with hazard ratios of 1.50 (95% CI: 1.30-1.73) in men and 1.25 (95% CI: 1.06-1.47) in women. These findings suggest that the association between handgrip strength and incident diabetes is stronger and clearer when handgrip strength is expressed relative to body size [19].

According to Byeon et al. [20], there was no significant association between AHGS and metabolic syndrome, but the odds ratio (OR) of metabolic syndrome prevalence was 2.52 times higher in men and 5.01 times higher in women in the low RHGS group compared with the high RHGS group. In that study, body weight is a confounder that may distort the interpretation of AHGS, and RHGS, which adjusts for body weight, is interpreted as more appropriate for explaining metabolic risk. This supports the findings of the present study. From the perspective of the metabolic load-capacity framework, RHGS can be interpreted as the ratio of muscular function (capacity) to body weight (load), and it shows a more stable and reproducible relationship with metabolic risk [12,21,22]. The fact that the effect size of RHGS was maintained even after adjusting for covariates in the present study supports this mechanism, and it also provides a basis for considering RHGS as a priority indicator for screening diabetes-related risk in clinical and epidemiological practice. However, since the choice of denominator (body weight vs. BMI) may affect how obesity is reflected, future sensitivity analyses with simultaneous adjustment for waist circumference and body fat percentage are needed to confirm the robustness of these findings.

In the present study, higher lower-body FFM was associated with lower odds of diabetes in the total sample. In sex-stratified analyses, this inverse association remained significant only among women. This suggests that the regional distribution of FFM (especially lower-body muscle), rather than the total amount, may be more sensitive in explaining metabolic health. Skeletal muscle is the main organ for insulin-stimulated glucose uptake [23]. Since the lower-body handles a large portion of the metabolic workload during walking and daily activities, sufficient lower-body muscle mass may be especially important in women. In fact, previous studies have reported that appendicular and leg FFM, or femoral muscle cross-sectional area, are closely associated with insulin resistance and type 2 diabetes, and segmental indicators tend to capture metabolic risk better than total FFM [8,17,24,25]. The pattern in which the signal was prominent only in women may be explained by sex differences in insulin sensitivity and fat and muscle distribution related to estrogen changes around menopause [26,27]. In men, the influence of other metabolic factors such as abdominal and hepatic fat may be relatively greater, and this may dilute the muscle mass signal [28]. Bias in FFM measurement should also be considered. The FFM used in the present study is an estimate based on bioelectrical impedance analysis (BIA), not a direct measurement, and its accuracy may depend largely on the prediction equations within the device and on standardized measurement conditions [29,30]. Because of the methodological characteristics of BIA, the results are sensitive to fluid status (dehydration, edema) and changes in the measurement environment, and differential measurement error is more likely to occur, especially in patients with diabetes [29,31]. In addition, total FFM is a comprehensive indicator that includes organs and total body water, so the metabolically meaningful skeletal muscle signal may be diluted [29,31]. Segmental FFM is sensitive to measurement posture and electrode contact conditions, so its variability may be increased [13]. Therefore, BIA-based results should be interpreted under standardized conditions, and although statistical significance was confirmed even after weight adjustment in the present study, cross-validation with reference methods such as dual-energy X-ray absorptiometry (DXA) is desirable to check estimation bias [30,32,33]. These findings suggest that lower-body muscle strengthening and improvement of muscular function should be systematically incorporated into clinical and public health strategies for diabetes risk assessment and prevention. In particular, prescribing and monitoring resistance and strength training targeted at women and at populations with low lower-body FFM may be useful.

The present study has several limitations. First, this is a cross-sectional study using KNHANES data, so the causal relationship between HGS or FFM and diabetes cannot be confirmed. Second, in addition to fasting blood glucose and HbA1c, self-reported items (physician diagnosis of diabetes, treatment status, and self-reported diabetes status) were used for diabetes diagnosis, so inaccurate diagnoses may have been included. Third, KNHANES does not distinguish between type 1 and type 2 diabetes, so disease heterogeneity may not be fully reflected in the present results. Fourth, although many covariates such as socioeconomic and lifestyle factors that may affect the results were controlled, not all confounders that may have affected the results were adjusted. Fifth, the FFM used in the present study is a BIA-based estimate, and compared with direct measurement methods, it may have measurement error and bias related to fluid status. Sixth, HGS is a convenient indicator that represents whole-body muscular function, but it has a limitation that it does not sufficiently reflect the function of specific muscle groups or qualitative muscle characteristics. Therefore, caution is required in generalizing the present results. Despite these limitations, the present study has significance in showing the importance of RHGS and lower-body FFM in relation to the prevalence of diabetes by examining a Korean adult population.

In this nationally representative sample of Korean adults aged 19 years or older, RHGS and lower-body FFM showed inverse associations with the odds of diabetes, whereas AHGS and total FFM showed less consistent associations. These findings suggest that muscle function- and region-specific muscle mass indicators may complement conventional anthropometric measures, such as body weight and BMI, in diabetes risk assessment.

In particular, RHGS may be a practical screening indicator because handgrip strength is inexpensive, simple to measure, and feasible for use in primary care and community-based health screening. Future longitudinal and intervention studies are needed to confirm whether improving muscle function and preserving lower-body muscle mass can reduce diabetes risk.