Comparison of nutrients and ultra-processed food consumption between different phenotypes defined by abdominal obesity and sarcopenia

TL;DR

This study compares nutrient and ultra-processed food consumption among adults with different health conditions like obesity and sarcopenia in rural Korea.

Contribution

The study identifies specific dietary patterns associated with sarcopenia and sarcopenic obesity using the NOVA classification system.

Findings

Higher protein intake was linked to lower odds of probable sarcopenia.

Ultra-processed food consumption was associated with increased odds of abdominal and sarcopenic obesity.

Minimally processed food intake was lower in the sarcopenic obesity group.

Abstract

Obesity, sarcopenia, and sarcopenic obesity require effective nutritional strategies in public health.

This cross-sectional study analyzed 535 adults aged 40–60 years in rural Korea. Participants were grouped by abdominal obesity (waist circumference ≥90 cm for men, ≥85 cm for women) and probable sarcopenia (handgrip strength ≤28 kg for men, ≤18 kg for women). Dietary intake was assessed using a semi-quantitative food frequency questionnaire and NOVA classification. Associations between food group intake and each phenotype were analyzed using multinomial logistic regression, adjusting for confounders.

Compared with healthy controls, the probable sarcopenia group reported lower intakes of total protein (p = 0.041), vitamin A (p = 0.041), and carotenoids (p = 0.046), and higher intake of processed culinary ingredients (p = 0.012). The sarcopenic obesity group had lower intake of minimally processed foods (p = 0.046) and higher intake of ultra-processed foods (p = 0.010). In regression models, higher protein intake was associated with lower odds of probable sarcopenia (OR = 0.73, 95% CI: 0.56–0.95). Higher ultra-processed food intake was associated with increased odds of abdominal obesity (OR = 1.26, 95% CI: 1.03–1.54) and sarcopenic obesity (OR = 1.37, 95% CI: 1.02–1.84).

Higher protein and minimally processed food intakes were associated with lower odds of probable sarcopenia, whereas higher ultra-processed food intake was associated with higher odds of abdominal and sarcopenic obesity in middle-aged adults in rural Korea. These observational findings may inform hypothesis generation and public health planning, but prospective studies are needed to confirm these associations and assess causality.

Tables

Table 1. General characteristics of the study participants by four phenotypes defined by abdominal obesity and probable sarcopenia.

Variable	Healthy controla	Central obesityb	Probable sarcopeniac	Sarcopenic obesityd	Total	p-value*
N (%)	239 (44.7)	176 (32.9)	70 (13.1)	50 (9.3)	535 (100.0)
Sex						0.013
Male	63 (26.4)	51 (29.0)	8 (11.4)	8 (16.0)	130 (24.3)
Female	176 (73.6)	125 (71.0)	62 (88.6)	42 (84.0)	405 (75.7)
Age						0.004
40–49	77 (32.2)	31 (17.6)	16 (22.9)	6 (12.0)	130 (24.3)
50–59	86 (36.0)	67 (38.1)	30 (42.9)	21 (42.0)	204 (38.1)
60–69	76 (31.8)	78 (44.3)	24 (34.3)	23 (46.0)	201 (37.6)
Marital status (N = 532)						0.751
Marriede	193 (81.1)	144 (82.8)	56 (80.0)	38 (76.0)	431 (81.0)
Not marriedf	45 (18.9)	30 (17.2)	14 (20.0)	12 (24.0)	101 (19.0)
Education level (N = 532)						<0.001
≤Middle school	52 (21.8)	67 (38.1)	18 (25.7)	20 (40.0)	157 (29.4)
High school	96 (40.2)	76 (43.2)	34 (48.6)	22 (44.0)	228 (42.6)
≥College	91 (38.1)	33 (18.8)	18 (25.7)	8 (16.0)	150 (28.0)
Household income per month (N = 532)						0.001
<2 million Won	50 (21.0)	52 (29.7)	14 (20.3)	22 (44.0)	138 (25.9)
2–4 million Won	85 (35.7)	59 (33.7)	34 (49.3)	19 (38.0)	197 (37.0)
>4 million Won	103 (43.3)	64 (36.6)	21 (30.4)	9 (18.0)	197 (37.0)
Smoking status (N = 529)						0.275
Never smokerg	184 (78.0)	130 (75.1)	59 (84.3)	39 (78.0)	412 (77.9)
Former smoker	21 (8.9)	27 (15.6)	6 (8.6)	6 (12.0)	60 (11.3)
Current smoker	31 (13.1)	16 (9.3)	5 (7.1)	5 (10.0)	57 (10.8)
Drinking status (N = 534)						0.290
Current drinker	161 (67.4)	122 (69.7)	43 (61.4)	24 (48.0)	350 (65.5)
Current abstainer	78 (32.6)	53 (30.3)	27 (38.6)	26 (52.0)	184 (34.5)
Recommended PA levels						0.650
No PA	131 (54.8)	97 (55.1)	43 (61.4)	33 (66.0)	304 (56.8)
Insufficient or inactive PA	37 (15.5)	32 (18.2)	11 (15.7)	5 (10.0)	85 (15.9)
Recommended PAh	71 (29.7)	47 (26.7)	16 (22.9)	12 (24.0)	146 (27.3)

Table 2. Daily nutrient intake by four phenotypes defined by abdominal obesity and probable sarcopenia.

Variable	Healthy controla	Central obesityb	Probable sarcopeniac	Sarcopenic obesityd	p-value
N	239	176	70	50
Total energy (kcal/day)	1,980.0 (55.3)	1,964.9 (65.1)	2,058.4 (102.4)	1,910.3 (118.0)	0.803
Carbohydrates (%kcal/day)	59.4 (0.7)	59.6 (0.9)	61.8 (1.4)	58.4 (1.6)	0.345
Total protein (%kcal/day)	14.7 (0.2)	14.5 (0.3)	13.7 (0.4)*	14.9 (0.5)	0.179
Fat (%kcal/day)	20.9 (0.5)	21.0 (0.6)	20.1 (0.9)	21.9 (1.1)	0.661
Saturated fatty acids (% kcal/day)	6.1 (0.2)	6.1 (0.2)	5.9 (0.3)	6.5 (0.3)	0.609
Monounsaturated fatty acids (%kcal/day)	6.6 (0.2)	6.6 (0.2)	6.4 (0.3)	7.0 (0.4)	0.670
Polyunsaturated fatty acids (%kcal/day)	5.4 (0.1)	5.5 (0.2)	5.0 (0.3)	5.4 (0.3)	0.427
Omega-3 fatty acids (%kcal/day)	0.6 (0.0)	0.7 (0.0)	0.6 (0.0)	0.6 (0.0)	0.252
Omega-6 fatty acids (%kcal/day)	4.9 (0.1)	5.0 (0.2)	4.5 (0.2)	4.9 (0.3)	0.429
Cholesterol (mg/day)	332.3 (12.2)	318.9 (14.4)	330.5 (22.6)	322.3 (26.1)	0.908
Dietary fiber (g/day)	21.7 (0.5)	22.7 (0.6)	20.9 (1.0)	21.2 (1.2)	0.400
Potassium (mg/day)	2,925.3 (62.8)	2,929.9 (73.9)	2,844.1 (116.4)	2,732.2 (134.1)	0.556
Calcium (mg/day)	487.2 (12.8)	476.8 (15.1)	447.0 (23.8)	485.1 (27.4)	0.514
P (mg/day)	1,060.8 (15.1)	1,044.5 (17.7)	1,018.4 (27.9)	1,043.2 (32.2)	0.596
Iron (mg/day)	14.6 (0.3)	14.7 (0.3)	13.6 (0.5)	14.0 (0.5)	0.153
Sodium (mg/day)	3,375.1 (85.8)	3,523.7 (101.0)	3,153.8 (159.1)	3,570.1 (183.3)	0.193
Vitamin A (μg RE/day)	673.6 (23.2)	661.4 (27.3)	567.3 (43.0)*	602.5 (49.6)	0.127
Retinol (μg /day)	104.9 (3.8)	97.4 (4.4)	95.7 (7.0)	100.3 (8.0)	0.519
Carotenoid (μg/day)	3,270.2 (130.4)	3,269.8 (153.5)	2,720.6 (241.8)*	2,906.1 (278.5)	0.154
Thiamine (mg/day)	1.9 (0.0)	1.9 (0.0)	1.9 (0.0)	1.9 (0.1)	0.819
Riboflavin (riboflavin, mg/day)	1.5 (0.0)	1.4 (0.0)	1.4 (0.1)	1.4 (0.1)	0.800
Niacin (mg/day)	14.0 (0.2)	13.7 (0.3)	13.1 (0.4)	13.6 (0.5)	0.346
Vitamin C (mg/day)	113.1 (5.7)	110.7 (6.7)	106.4 (10.6)	89.2 (12.2)	0.363

Table 3. Intake of NOVA food groups (MPF, PCI, PF, UPF) by four phenotypes defined by abdominal obesity and probable sarcopenia.

	Healthy controla	Central obesityb	Probable sarcopeniac	Sarcopenic obesityd	p-value
MPF (NOVA 1)	53.8 (1.1)	52.5 (1.3)	55.6 (2.0)	48.7 (2.3)*	0.116
PI (NOVA 2)	0.54 (0.1)	0.39 (0.1)	1.09 (0.2)*	0.70 (0.2)	0.021
PF (NOVA 3)	27.29 (0.7)	26.66 (0.9)	24.18 (1.4)*	27.46 (1.6)	0.236
UPF (NOVA 4)	18.4 (0.8)	20.5 (0.9)	19.1 (1.5)	23.2 (1.7)*	0.051

Table 4. Intakes of nutrients and food groups and their odds ratios for central obesity and probable sarcopenia status using multinomial logistic regression models.

	Total protein	Vitamin A	Carotenoids	MPF (NOVA 1)	PCI (NOVA 2)	PF (NOVA 3)	UPF (NOVA 4)
Healthy controla	1.00 (REF)	1.00 (REF)	1.00 (REF)	1.00 (REF)	1.00 (REF)	1.00 (REF)	1.00 (REF)
Central obesityb	0.95 (0.78–1.16)	1.05 (0.86–1.27)	1.05 (0.87–1.28)	0.92 (0.77–1.12)	0.92 (0.79–1.1.08)	0.98 (0.81–1.19)	1.26 (1.03–1.54)
Probable sarcopeniac	0.73 (0.56–0.95)	0.85 (0.66–1.11)	0.85 (0.65–1.10)	1.11 (0.86–1.44)	1.16 (0.94–1.43)	0.77 (0.59–0.99)	1.09 (0.83–1.42)
Sarcopenic obesityd	1.07 (0.79–1.44)	0.86 (0.64–1.15)	0.84 (0.63–1.13)	0.75 (0.56–1.01)	1.05 (0.84–1.34)	1.03 (0.77–1.38)	1.37 (1.02–1.84)

Full text

Introduction

1

The prevalence of abdominal obesity has shown a steady increase across all age groups in Korea, reaching 38.4% in 2021 (1). This trend is particularly concerning as abdominal obesity is strongly associated with an elevated risk of various chronic diseases. Korean studies have reported significant links between abdominal obesity and the incidence of hypertension, dyslipidemia, and diabetes in middle-aged and older adults (2). Additionally, individuals with abdominal obesity have a higher relative risk of developing type 2 diabetes, myocardial infarction, and certain cancers compared to those with normal weight (3), and older adults with abdominal obesity are more likely to experience frailty (4).

Changes in body composition, characterized by increased fat mass and decreased muscle mass, are an inherent aspect of the aging process and significantly impact the health of older adults (5, 6). Sarcopenia, an age-related condition characterized by progressive loss of muscle mass, decreased muscle strength, and impaired muscle function, poses a significant public health concern among middle-aged and elderly populations (7). Research has consistently demonstrated the association of sarcopenia with various health conditions, including an increased prevalence of metabolic syndrome (8), functional impairment and physical disability (9), increased risk of falls and fractures (10), as well as a higher risk of frailty and mortality (10, 11).

Sarcopenic obesity, characterized by low muscle mass, muscle quality, or strength combined with high fat mass, has emerged as a major health concern among older adults, representing a new form of obesity in the elderly population (12). Due to the increased metabolic burden caused by both sarcopenia and obesity, sarcopenic obesity plays a significant role in cardiometabolic risk (13). Research has demonstrated that individuals with sarcopenic obesity are at a higher risk of cardiovascular diseases (14) and mortality compared to individuals without sarcopenia (15). Furthermore, individuals with sarcopenic obesity were found to have a higher risk of cardiovascular diseases, diabetes, and physical dysfunction compared to those with either sarcopenia or obesity alone (16). Both sarcopenia and sarcopenic obesity are key factors that hinder healthy aging and are closely linked to quality of life (17, 18).

Obesity, sarcopenia, and sarcopenic obesity have emerged as major health concerns in modern society, with dietary factors playing a crucial role in their development and progression. Existing research suggests that obesity results from an imbalance between energy intake and expenditure (19), while inadequate protein intake and nutritional imbalances are recognized as primary contributors to sarcopenia and sarcopenic obesity. A high-protein diet is particularly essential for maintaining muscle mass and function, with substantial research focused on the impact of protein intake in sarcopenia prevention. Additionally, specific nutrients, such as vitamin D and omega-3 fatty acids, have received considerable attention for their effects on muscle health and body fat regulation (20).

Recently, dietary patterns based on the level of food processing have gained attention. The NOVA classification categorizes foods into minimally processed foods (MPF, NOVA1), processed culinary ingredients (PCI, NOVA2), processed foods (PF, NOVA3), and ultra-processed foods (UPF, NOVA4) (21). While MPF intake is generally associated with better dietary quality (21) high consumption of PCI, such as salt and sugar, can indirectly reduce dietary quality by displacing nutrient-dense foods (22, 23). PF may exert neutral or beneficial metabolic effects depending on their composition, as observed with certain items like cheese (24). In contrast, high UPF intake has been linked to increased risks of frailty (25), decline in grip strength (26), and muscle mass loss (27). However, no studies have comprehensively investigated associations between intakes of all four NOVA groups and obesity and sarcopenia among middle-aged and older Korean adults. Therefore, dietary factors related to obesity, sarcopenia, and sarcopenic obesity should be examined more comprehensively by considering the level of food processing.

Sarcopenia is associated with increased risks of falls, fractures, functional decline, and mortality, and is often left undiagnosed, highlighting the importance of early detection and prevention (28, 29). Probable sarcopenia (PS), defined by reduced handgrip strength without assessments of muscle mass or physical performance, has recently gained attention as a practical early marker of sarcopenia risk. According to the European Working Group on Sarcopenia in Older People (EWGSOP), low muscle strength is the most critical determinant of sarcopenia, and early intervention is essential once strength decline is observed (30). Although most sarcopenia studies have focused on older adults aged 60 and over, muscle strength tends to decline from midlife. Therefore, studies that include middle-aged adults are essential for informing early prevention strategies. PS is particularly suitable for assessing sarcopenia risk in both middle-aged and older populations and may serve as an effective tool for early identification and management.

Gangwon Province in South Korea is a super-aged region where more than 20 percent of the population is aged 65 years or older, and its terrain is largely mountainous and rural (31). According to the 2022 Community Health Survey, the province has the second-highest obesity rate in the country (32). In response, the Gangwon Obesity and Metabolic Syndrome (GOMS) study was launched to investigate the factors contributing to the high obesity prevalence and to support region-specific public health strategies (33). Using data from the GOMS cohort, the present study aimed to assess the prevalence of abdominal obesity and PS in adults aged 40 to 60 years and to examine associated dietary factors, with the goal of providing evidence to support early prevention efforts targeting rural populations.

Methods

2

Study design and participants

2.1

This was a cross-sectional study utilizing baseline survey data from the GOMS cohort. The GOMS cohort is an ongoing community-based prospective cohort study designed to assess body weight and body composition status among middle-aged and older adults residing in rural areas of the Gangwon Province, investigating obesity-related factors, and examining longitudinal changes in obesity (33). The study participants consisted of Korean residents aged 20–69 years living in the Yeongseo region of Gangwon Province. The exclusion criteria comprised illiterate individuals, those with moderate to severe cognitive impairment, and those currently participating in other observational or clinical studies. The first-year baseline survey was conducted from June to July 2022, recruiting 319 articipants, while the second survey was conducted from September to October 2023, recruiting 317 participants. After excluding individuals with missing key measurements, such as grip strength, a total of 535 participants (130 males and 405 females) aged 40–60 was included in the final analysis. This study was approved by the Institutional Review Board of the authors’ institution.

Data collection procedures

2.2

The data collection process was conducted by trained researchers and research assistants. Before data collection, experienced researchers conducted two training sessions, comprising both theoretical instruction and hands-on practice. The training covered all aspects of the survey, including direct practice in anthropometric measurements and grip strength assessments. To minimize measurement errors, simulations were employed for challenging measurements. Additionally, study participants received presurvey materials to facilitate their understanding of the survey procedures.

Upon arrival at the designated survey site, participants received a detailed explanation of the study’s purpose and procedures, followed by the obtention of written informed consent. Data collection included anthropometric measurements and grip strength assessments. Additionally, participants completed a self-administered questionnaire, which gathered information on demographic characteristics, physical activity levels, alcohol consumption habits, smoking status, and dietary intake patterns. To ensure data accuracy, trained researchers conducted follow-up interviews to review the completed questionnaires, addressing any missing responses or correcting inaccurate answers.

Dietary assessment

2.3

Includes macronutrient and micronutrient intake, and NOVA-based food classification

2.3.1

Dietary intake was assessed using a semiquantitative Food Frequency Questionnaire (FFQ), which was developed and validated through the Korea National Health and Nutrition Examination Survey (KNHANES) to evaluate the habitual dietary intake of Korean adults (34, 35). The FFQ consists of 112 food items, and participants reported their consumption frequency and portion size for each item over the past year. The frequency categories were divided into nine levels (never or rarely, once per month, two to three times per month, once per week, two to four times per week, five to six times per week, once per day, twice per day, and three times per day). Portion sizes were categorized as small, medium, and large (34). Participants with daily caloric intakes outside the plausible range (less than 500 kcal or more than 5,000 kcal) were excluded from the study to ensure data validity.

Dietary data obtained from the FFQ were categorized using the NOVA food classification system, which groups foods according to the extent and purpose of industrial processing (36). Because the NOVA system was originally developed based on Brazilian dietary patterns, its direct application to Korean mixed dishes is limited. Therefore, this study referred to the Korean-adapted NOVA classification framework developed to reflect the Korean food environment (37) and applied the FFQ-based NOVA classification method validated for use with FFQ data (38). This method demonstrated good agreement between FFQ-based and 24-h recall-based estimates of UPF intake using data from the KNHANES 2016, confirming the applicability of the NOVA classification to FFQ data in Korean adults. Based on this validated framework, 109 FFQ items (excluding three alcohol-related items) were classified into the four NOVA groups according to the item-specific categorization matrix presented by Jung et al. (38). The contribution of each NOVA group to total energy intake was calculated as a percentage of total daily caloric intake. Detailed classification of all FFQ items into the four NOVA groups is provided in Supplementary Table S2.

Anthropometric and muscle strength measurement

2.4

Includes abdominal obesity and muscle strength assessments

2.4.1

Body mass index (BMI) has limitations in assessing obesity in older adults, as it does not account for age-related muscle loss (39). In contrast, abdominal obesity, measured by waist circumference (WC), is a more sensitive indicator of visceral fat accumulation and is a better predictor of the risk of metabolic syndrome and cardiovascular diseases in middle-aged and older adults (40, 41). In this study, trained examiners measured WC to the nearest 0.1 cm using a measuring tape at the midpoint between the lower rib margin and the iliac crest. Abdominal obesity was defined according to the Korean Society for the Study of Obesity (KSSO) criteria, with cutoff values of ≥90 cm for males and ≥85 cm for females (42).

The diagnosis of sarcopenia typically involves a comprehensive assessment of muscle strength, muscle mass, and physical function. This study focused on middle-aged and older adults aged 40–60, emphasizing the importance of early detection and preventive intervention for sarcopenia. To identify individuals at risk, the Asian Working Group for Sarcopenia (AWGS) criteria for PS were applied (43). PS is a method for screening sarcopenia risk solely based on muscle strength, without evaluating muscle mass or physical function. According to AWGS guidelines, PS is defined as a grip strength of ≤28 kg for males and ≤18 kg for females (43). Grip strength was measured using a digital hand dynamometer (TKK-5401; TAKEI, Tokyo, Japan) following standardized procedures. Participants were instructed to stand upright with their wrist and arm in a neutral position, and their elbow fully extended. Grip strength was measured once on each hand, and the average value was used as the final muscle strength score. To ensure consistency in measurement conditions, all participants received detailed instructions on proper posture and exerting maximum effort before the assessment.

Phenotype classification and statistical analysis

2.5

Defines phenotypes and details statistical methods used

2.5.1

Participants were classified into four groups based on the AWGS-defined PS criteria and the KSSO-defined abdominal obesity criteria. The healthy control group consisted of individuals who did not meet the criteria for either PS or abdominal obesity. The obesity group comprised participants who had abdominal obesity but did not meet the PS criteria. The PS group comprised individuals who met the PS criteria but did not have abdominal obesity. Lastly, the sarcopenic obesity group consisted of participants who met both the PS and abdominal obesity criteria. For the purpose of this study, the term sarcopenia is used interchangeably with PS, as sarcopenic obesity is defined as the coexistence of PS and abdominal obesity. By classifying participants based on abdominal obesity and PS status, this study aimed to identify the risk stages of sarcopenia, focusing on muscle weakness, and conduct an in-depth analysis of its interaction with abdominal obesity.

Covariates

2.5.2

The key covariates comprised biological (gender, age), socioeconomic (education level, household income), and lifestyle domains (drinking, smoking, physical activity). The main covariates included age, gender, education level, household income, drinking, smoking, and physical activity. Age was used as a continuous variable, and education level was classified into middle school graduation or lower, high school graduation, and university graduation or higher. Household income was classified into less than 2 million Korean won, 2 to 4 million won, and more than 4 million won based on monthly household gross income. Drinking was defined as the current drinking status (yes or no), and smoking was classified into non-smoking (less than 100 cigarettes in a lifetime), past smoking, and current smoking. Physical activity was classified into three categories: no physical activity, insufficient or inactive physical activity, and recommended physical activity. Recommended physical activity was defined according to standard guidelines as engaging in at least 150 min of moderate-intensity activity or 75 min of vigorous-intensity activity per week (44). In addition, minutes of vigorous-intensity activity were counted as double when combined with moderate-intensity activity. Based on this, whether physical activity was satisfied was classified into no physical activity, not meeting physical activity, and met physical activity.

Statistical analysis

2.5.3

Descriptive statistical analysis was performed to present the general characteristics of the subjects and the ratio or average of the outcome variables. In addition, the relationship between abdominal obesity and sarcopenia and dietary factors was analyzed after correcting for major covariates using multiple regression analysis. To compare differences between groups, dummy variables were created for group classification, and multiple regression analysis was performed with the healthy control group as the reference. Post-estimation analyses, including hypothesis tests and linear combination tests, were conducted to assess statistical differences between the healthy control group and other groups. Additionally, multinomial logistic regression models were used to estimate covariate-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for abdominal obesity, PS, and sarcopenic obesity, using the healthy control group as the reference. Quartiles 2, 3, and 4 of each dietary factor were compared with quartile 1. All analyses were performed using Stata MP 17.0 (Stata Corp LLC, College Station, TX, USA).

Results

3

The general characteristics of participants by obesity and PS status are presented in Table 1. The mean age was 56.0 years, and the proportion of females was higher (75.7%, p = 0.013). The age distribution was 24.3% aged 40–49 years, 38.1% aged 50–59 years, and 37.6% aged 60–69 years (p = 0.004). The PS group showed a higher proportion of individuals in their 50s, whereas the obesity and sarcopenic obesity groups showed relatively higher proportions of individuals in their 60s. The obesity and sarcopenic obesity groups showed the lowest proportion with a college education or higher. The sarcopenic obesity group showed a relatively higher proportion with income <2 million KRW. No statistically significant between-group differences were observed for smoking status, alcohol consumption, or physical activity.

Table 2 summarizes differences in daily macro- and micronutrient intake by obesity and PS status. Compared with the healthy control group, the PS group showed lower mean intakes of total protein (p = 0.041), vitamin A (p = 0.041), and carotenoids (p = 0.046). No other nutrients differed significantly.

Table 3 presents adjusted mean intakes of NOVA food groups by obesity and PS status. Relative to healthy controls, the PS group showed higher intake of PCI (p = 0.012) and lower intake of PF (p = 0.047). The sarcopenic obesity group showed lower intake of MPF (p = 0.046) and higher intake of UPF (p = 0.010).

Table 4 shows associations of food intake with odds across groups defined by obesity and PS status. Per-quartile higher total protein intake was associated with lower odds of PS (OR = 0.73, 95% CI: 0.56–0.95), and per-quartile higher PF intake was similarly associated with lower odds of PS (OR = 0.77, 95% CI: 0.59–0.99). In contrast, per-quartile higher UPF intake was associated with higher odds of abdominal obesity (OR = 1.26, 95% CI: 1.03–1.54) and sarcopenic obesity (OR = 1.37, 95% CI: 1.02–1.84). Higher MPF intake (per quartile) showed a borderline association with lower odds of sarcopenic obesity (OR = 0.75, 95% CI: 0.56–1.01).

Discussion

4

This cross-sectional study examined associations between dietary intake and four groups defined by obesity and PS status among adults aged 40–60 years in Gangwon Province. Compared with the healthy control group, the PS group showed lower adjusted mean intakes of protein, vitamin A, and carotenoids, whereas the sarcopenic obesity group showed lower intake of MPF and higher intake of UPF.

In this study, the prevalence of PS and sarcopenic obesity was 13.1% and 9.3%, respectively. A Korean study applying the same criteria reported lower prevalence (PS 8.6%, sarcopenic obesity 6.2%) among adults aged ≥40 years (45), and a UK cohort with an older mean age (76 years) reported PS prevalence ranges of 7.7%–21.1% in men and 5.9%–21.3% in women (46). Despite the younger mean age in our sample (56.0 years), the observed prevalences were comparatively high. This pattern may reflect the rural context of the study population. Previous studies have reported higher prevalence of obesity and PS, as well as less favorable muscle health indicators, in rural versus urban areas (47–49), including higher PS prevalence among rural residents in Korea and India (45, 50).

The proportion of energy from protein was significantly lower in the PS group (13.8%) than in the healthy control (14.7%), obesity (14.6%), and sarcopenic obesity (14.9%) groups. Per-quartile higher total protein intake was associated with lower odds of PS. The average protein energy ratio among participants (14.3%) was below the 2019 national average for Korean adults (15.6%) (42), while remaining within the recommended range of 7%–20% (51). Notably, protein intake in the PS group was lower despite already modest intake at the population level. These findings are consistent with prior reports of associations between higher protein intake and more favorable muscle mass and strength in middle-aged and older adults (52), as well as Korean data reporting associations between lower protein intake, reduced grip strength, and higher PS risk (53).

In this study, the PS group had significantly lower intakes of vitamin A and carotenoids compared with the healthy control group, which is consistent with prior reports on their potential roles in muscle health. Vitamin A’s antioxidant and anti-inflammatory properties may influence chronic inflammation in muscle and adipose tissue, which could be relevant to muscle maintenance (54). Several studies have reported associations between lower circulating carotenoid levels and faster decline in walking speed (55), greater risk of frailty (56), and accelerated loss of muscle strength (57), particularly among females. In the present study, carotenoid intake remained lower in the PS group after adjustment for sex and other covariates, a pattern compatible with a potential role of carotenoids in muscle health across sexes. Given the structural diversity of carotenoids and vitamin A, further work using prospective designs and biomarker-based assessments is needed to clarify compound-specific mechanisms and interactions with protein, particularly in populations with relatively low protein intake, and to strengthen causal inference.

This study found no significant associations between sarcopenia and other macronutrients or micronutrients aside from protein, carotenoids, and vitamin A. Although previous research suggested roles for vitamin D, vitamin C, and omega-3 fatty acids in sarcopenia prevention, findings remain inconsistent or inconclusive (20), potentially due to regional and lifestyle differences. Traditional Korean diets common in rural areas, characterized by high carbohydrate intake from refined grains (58, 59), may be associated with relatively lower protein intake. Such dietary patterns, shared across groups, may partly explain the lack of nutrient intake differences observed between sarcopenia and control groups. Future studies should incorporate biochemical markers to objectively evaluate nutrient status and clarify the complex relationships between diet and muscle health.

In this study, there were significant differences in PCI and PF intake between the healthy control and PS groups. Notably, higher PF intake was associated with lower odds of PS, whereas PCI differed in the opposite direction. PCI primarily includes culinary ingredients such as oils, salt, and sugar, which are rarely consumed independently but shape overall diets through meal preparation and seasoning (36). This interpretation is consistent with prior evidence showing that higher intakes of added sugars and sodium, core PCI components, are associated with lower overall diet quality in adult populations (60, 61). The Brazilian Dietary Guidelines, informed by the NOVA classification system, specifically recommend using PCI in small amounts and in moderation during culinary preparations, as excessive use can render diets nutritionally unbalanced (62). In the Korean context, PF includes traditionally processed fermented foods such as kimchi, which may have neutral or beneficial nutritional effects depending on other dietary components. Specifically, traditional Korean diets rich in fermented foods have been shown to exert beneficial effects on metabolic health, including improvements in blood pressure and glycemic control (63). However, these observations likely reflect regional dietary characteristics and should not be generalized to processed foods as a whole. Although we adjusted for multiple covariates, the cross-sectional design warrants cautious interpretation, and longitudinal studies are needed. Moreover, evidence directly linking PCI or PF intake to muscle-related outcomes remains limited, and further research examining the associations between different NOVA processing categories and muscle health is needed.

Participants with sarcopenic obesity reported higher UPF consumption, consistent with previous findings showing greater UPF intake among individuals with sarcopenic or obese phenotypes (27). However, longitudinal studies examining UPF intake and muscle outcomes have shown inconsistent results. A prospective Chinese cohort found that higher UPF consumption was associated with faster decline in handgrip strength over 3 years (26). Recent evidence from the Framingham Offspring Study also demonstrated that higher UPF intake was associated with increased odds of frailty development and annual declines in grip strength and gait speed over a 10-year follow-up period (64). Despite these findings, a systematic review and meta-analysis reported inconsistent evidence, with cohort studies showing significant associations with frailty but not with muscle strength, while cross-sectional studies indicated associations with low muscle strength (65). Given the cross-sectional design of the present study, reverse causation cannot be ruled out; individuals with sarcopenic obesity may have a greater tendency to choose convenient or ultra-processed foods due to functional limitations or reduced cooking capacity.

In South Korea, the contribution of UPFs to total food intake increased from 23.1% in 2010–2012 to 26.1% in 2016–2018, showing a consistent upward trend across all age and socioeconomic groups (66). In the present study, UPFs accounted for 23.2% of total food intake, comparable to the national level in 2010–2012. Although no significant differences in individual nutrient intakes were observed, the contrasting patterns of MPF and UPF consumption may reflect differences in overall dietary quality related to muscle health. Given the cross-sectional design and the growing public health relevance of UPF consumption, additional prospective studies are warranted to clarify the relationship between UPF intake and muscle-related outcomes.

This study has several limitations that should be considered when interpreting the findings. First, the cross-sectional design does not allow causal inference, and reverse causation cannot be completely ruled out. Second, although this study adjusted for multiple sociodemographic and lifestyle covariates, the possibility of residual confounding cannot be excluded. Third, dietary intake assessed via a self-reported food frequency questionnaire may be subject to recall bias, and the NOVA classification adapted to Korean mixed dishes may not be fully validated. Future studies incorporating objective biomarkers of dietary intake would help validate these findings and reduce measurement error. Fourth, although the overall sample size was adequate for primary analyses, statistical power may have been limited for detecting certain associations in subgroup analyses. Fifth, selection bias may have occurred as participants were community-dwelling volunteers who may differ from non-participants in health status and dietary behaviors, and survival bias cannot be ruled out given that individuals with severe sarcopenia or poor health may have been less likely to participate. Finally, as participants were recruited from rural and mountainous areas in Gangwon Province, the findings may not be generalizable to urban populations or those with different cultural and dietary contexts.

Despite these limitations, this study has several notable strengths. To our knowledge, few studies have examined the relationship between sarcopenic obesity phenotypes and NOVA food processing levels, particularly UPF consumption, among Korean adults aged under 70 years. Using PS instead of clinically defined sarcopenia is meaningful, as early identification of individuals at risk before the onset of clinically diagnosed sarcopenia is essential for prevention and public health interventions. Because the participants were community-dwelling adults under 70 years who generally maintain normal physical function, evaluating PS provides a more appropriate perspective on early metabolic and muscular risk profiles. Taken together, despite its cross-sectional nature, this study provides valuable evidence for understanding the role of dietary processing level in muscle health and may inform future prevention strategies targeting middle-aged and older adult populations.

In this cross-sectional study, lower intakes of protein, vitamin A, and carotenoids, together with higher PCI intake, were associated with higher odds of PS, whereas higher PF intake was associated with lower odds of PS. Higher UPF consumption was linked to greater odds of obesity and sarcopenic obesity, while higher MPF intake showed an inverse association with sarcopenic obesity. These findings provide preliminary evidence that differences in food processing levels may be related to early metabolic and muscle health profiles. Further longitudinal research is needed to confirm these associations and clarify their implications for prevention strategies.