Association between menopausal status and the probability of steatotic liver disease in middle-aged Korean women: a cross-sectional study using the Korea National Health and Nutrition Examination Survey 2007–2023
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Metabolic dysfunction-associated steatotic liver disease (SLD) emphasizes that metabolic dysfunction is a key driver of hepatic steatosis. Although menopause is characterized by substantial hormonal and metabolic shifts, population-based evidence on its association with the probability of SLD under the updated diagnostic framework remains limited. Therefore, this study investigated the association between menopausal status and probability of SLD in middle-aged Korean women.
Methods
We analyzed data of 15,106 women aged 40 to 59 years who participated in the Korea National Health and Nutrition Examination Survey (2007–2023). The probability of SLD was assessed using the hepatic steatosis index (≥36) in the presence of at least one metabolic abnormality. Menopausal status was self-reported. Weighted multivariate logistic regression was performed after adjusting for age, body mass index, waist circumference, and major metabolic risk factors. Effect modification was evaluated through subgroup analysis.
Results
Among the 15,106 women, 31.5% were postmenopausal and 22.1% met the criteria for SLD probability. Post-menopausal women exhibited higher metabolic risk profiles and a higher probability of SLD than premenopausal women. Menopausal status remained independently associated with SLD probability after multivariable adjustment (adjusted odds ratios, 1.22; P=0.035). The association varied by age; it was strongest among women aged 40 to 49 years and progressively attenuated in older age groups (P for interaction=0.001).
Conclusion
Menopausal status is independently associated with a higher probability of SLD among middle-aged Korean women, an association most pronounced in the early midlife. These findings highlight the importance of proactive metabolic screening and timely intervention during this critical period.
Steatotic liver disease (SLD) has emerged as one of the most pressing global health challenges of the 21st century. Affecting more than one-third of the world’s adult population, it is the most common cause of chronic liver disease [1]. Although frequently asymptomatic in its early stages, SLD may progress to steatohepatitis, advanced fibrosis, cirrhosis, and hepatocellular carcinoma, posing a substantial burden on healthcare systems worldwide [2-4].
The evolution from the traditional nonalcoholic fatty liver disease (NAFLD) nomenclature to SLD represents a fundamental paradigm shift in understanding the pathophysiology of fatty liver disease. Expert consensus has redefined the terminology, introducing SLD as an overarching concept that encompasses metabolic dysfunction-associated steatotic liver disease (MASLD), metabolic dysfunction, alcohol-associated liver disease (MetALD), and alcohol-related liver disease (ALD) [5,6]. In contrast to the exclusionary criteria-based definition of NAFLD, MASLD emphasizes the presence of metabolic dysfunction as the primary driver of hepatic steatosis, providing a more mechanistically oriented diagnostic framework. This refined definition improves diagnostic precision and facilitates targeted therapeutic approaches by highlighting the central role of metabolic abnormalities in disease pathogenesis [7].
Recent meta-analyses have estimated the SLD global prevalence at approximately 37.5%, with MASLD being the dominant subtype (33.6%), followed by MetALD (4.1%) and ALD (2.2%) [1]. In South Korea, a nationwide systematic review and meta-analysis reported that NAFLD affected approximately 30% of the adult population, with a substantially higher prevalence in men than women (41.1% vs. 20.3%) [8]. Although these estimates were derived using the traditional NAFLD definition, they may reflect a considerable population-level burden of metabolically driven SLD under the updated SLD framework. Previous epidemiological studies have established strong associations between MASLD and various demographic and metabolic factors, including age, obesity, diabetes mellitus, dyslipidemia, and hypertension, which have been consistently identified as major risk factors for MASLD development and progression [9,10]. These findings highlight the strong association between SLD and metabolic dysfunction, and reinforce the concept that SLD is a systemic metabolic disorder rather than a liver-limited condition.
Beyond conventional metabolic risk factors, increasing attention has been directed toward hormonal influences on SLD susceptibility, particularly regarding sex differences in disease prevalence and severity. Menopause is a critical transitional period in women’s lives, characterized by profound hormonal changes that extend beyond reproductive function. The decline in estrogen levels associated with menopause triggers a cascade of metabolic alterations, including increased visceral adiposity, insulin resistance, dyslipidemia, and elevated blood pressure, all known contributors to SLD [11,12]. Previous studies have demonstrated an association between menopause and NAFLD, and between NAFLD and metabolic abnormalities [13,14]. However, large-scale population-based studies examining menopause in relation to the probability of SLD within an updated disease framework remain limited.
Therefore, this study aimed to investigate the association between menopausal status and the probability of SLD in a large, nationally representative sample of Korean women aged 40 to 59 years, using data from the Korea National Health and Nutrition Examination Survey (KNHANES) conducted between 2007 and 2023. By clarifying this relationship, we sought to evaluate whether menopause was independently associated with a higher probability of SLD compared to established metabolic risk factors.
Methods
Study population
This cross-sectional study utilized data from the KNHANES conducted between 2007 and 2023. The KNHANES is a nationally representative survey conducted by the Korea Centers for Disease Control and Prevention that employs a stratified, multistage probability sampling design to assess the health and nutritional status of the Korean population. Women aged 40 to 59 years were included in the analysis [15], which was selected to encompass the Korean average age at menopause, reported as 49.4 years (standard deviation, ±5.1 years), by including women within approximately two standard deviations above and below the mean [15]. Exclusion criteria included: women who underwent hysterectomy (n=301); those currently using hormone replacement therapy or estrogen supplements (n=311); those with incomplete data on key variables (n=5,556); those who reported a fasting time of less than 8 hours (n=364); and those with other causes of SLD (n=15). The final analytical sample comprised 15,106 women (Figure 1).
Definition of steatotic liver disease
In this study, the probability of SLD was assessed in accordance with the updated nomenclature, using a surrogate index to reflect hepatic steatosis in combination with at least one metabolic risk factor. Hepatic steatosis was estimated using the hepatic steatosis index (HSI), calculated as follows: HSI=8×(alanine aminotransferase [ALT]/aspartate aminotransferase [AST] ratio)+body mass index (BMI) (+2 for the presence of diabetes mellitus). An HSI value of ≥36 was used to indicate a high probability of hepatic steatosis, consistent with validation studies conducted in Korean populations [16]. Additionally, at least one of the following metabolic risk factors was required to identify individuals with an increased probability of SLD: (1) overweight/obesity, defined as BMI ≥23 kg/m2, or waist circumference ≥85 cm for Asian women [17]; (2) elevated fasting serum glucose (≥100 mg/dL) or hemoglobin A1c (HbA1c) ≥5.7%, or a diagnosis of or treatment for diabetes mellitus; (3) hypertension (blood pressure ≥130/85 mm Hg or use of antihypertensive medications); or (4) dyslipidemia, plasma triglycerides ≥150 mg/dL, high-density lipoprotein cholesterol (HDL-C) <50 mg/dL, or use of lipid-lowering medication [18]. Accordingly, the SLD classification in this study reflected the probability of SLD based on validated surrogate indices, rather than a definitive clinical diagnosis.
Assessment of menopausal status
Menopausal status was determined based on self-reported responses in KNHANES. Women were categorized as either premenopausal or postmenopausal according to their menstrual history.
Covariate assessment
Potential confounding variables were included in the analyses based on their clinical relevance and previous studies. These variables included age, BMI, waist circumference, smoking status, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia. Hypertension was defined as taking antihypertensive medications or having a blood pressure of 140/90 mm Hg or higher on the day of the examination. Diabetes mellitus was defined as taking diabetes medications or having a fasting blood glucose of 126 mg/dL or higher, or an HbA1c of 6.5 or higher. Dyslipidemia was defined as taking medication or having a low-density cholesterol level of 130 mg/dL.
Laboratory measurements included fasting glucose, HbA1c, lipid profiles (total cholesterol, triglycerides, HDL-C, and low-density lipoprotein cholesterol [LDL-C]), and liver enzyme levels (ALT and AST). All blood samples were collected after at least 8 hours of fasting.
Lifestyle factors were assessed using a questionnaire. Regular physical activity was defined as at least 150 min/wk of moderate-intensity aerobic activity, 75 min/wk of vigorous-intensity aerobic activity, or an equivalent combination. Smoking status was categorized as never smoked, former smoker, or current smoker. Alcohol consumption was assessed according to the frequency and amount of alcohol consumed. Participants were categorized into three groups: non-drinkers or light drinkers (<14 drinks/wk), moderate drinkers (14–35 drinks/wk), and heavy drinkers (>35 drinks/wk).
Statistical analysis
Statistical analyses were performed using STATA ver. 18.0 (StataCorp LLC). To account for the complex survey design of the KNHANES and ensure nationally representative estimates, sampling weights were applied to all inferential analyses.
Baseline characteristics are presented using unweighted descriptive statistics to compare the characteristics between premenopausal and postmenopausal women (Table 1). Categorical variables were compared using chi-square tests; and continuous variables were compared using independent t-tests. Continuous variables, including age, BMI, and waist circumference, are presented as mean±standard deviation; while categorical variables, including smoking status, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia, are presented as numbers (%).
All primary inferences regarding the association between menopausal status and probability of SLD were derived from weighted multivariate analyses (Tables 2, 3), in which sampling weights provided by the KNHANES were applied to the complex survey design. The KNHANES weights were constructed through a multistage process, including design weights (based on selection probabilities), nonresponse adjustment weights, and post-stratification weights calibrated to Korean population demographics by region, sex, and age. Weight-trimming procedures were applied to address extreme values, followed by recalibration to maintain population representativeness. These weights were corrected for unequal selection probabilities, sampling differences, and nonresponse bias, ensuring nationally representative and unbiased estimates in this study [19].
Weighted multivariate logistic regression models evaluated the association between menopausal status and probability of SLD. Three models were constructed: (1) an unadjusted model; (2) model 1, adjusted for controlling for age, BMI, waist circumference, smoking status, alcohol consumption, and physical activity; and (3) model 2, adjusted for hypertension, diabetes mellitus, and dyslipidemia. The results were presented as odds ratios (ORs) with 95% confidence intervals (CIs). Statistical significance was set at P<0.05.
In additional analyses, menopausal duration was calculated as the difference between current age and age at menopause, and was included as an additional covariate to assess whether the association between menopausal status and SLD probability was independent of time since menopause.
Subgroup analyses examined the potential effect modifications of age, BMI, waist circumference (analyzed as continuous variables); smoking status, physical activity, alcohol consumption, hypertension, diabetes mellitus, and dyslipidemia (analyzed as categorical variables). Interaction terms between menopausal status and each covariate were tested, and P-values for interactions were reported.
Ethics statement
This study received exemption approval from the Institutional Review Board of Seoul National University Hospital (approval no., E-2308-103-1459). Obtaining informed consent was considered unnecessary because the participants’ agreement had already been secured through KNHANES. The datasets available in the public domain lacked specific personal identification details.
Results
Baseline characteristics
Among the 15,106 women aged 40 to 59 years included in the analysis, 10,346 (68.5%) were premenopausal and 4,760 (31.5%) were postmenopausal. The overall probability-based prevalence of SLD was 22.1% (n=3,339), with a significantly higher proportion observed among postmenopausal women (24.96% vs. 20.79%, P<0.001).
Baseline characteristics are summarized in Table 1. Postmenopausal women were older (mean age, 54.67 years vs. 47.48 years; P<0.001), and had higher BMI, waist circumference, systolic and diastolic blood pressure, fasting glucose, HbA1c, total cholesterol, triglycerides, LDL-C levels, ALT, and AST compared to premenopausal women (P<0.001). However, HDL-C levels and alcohol consumption patterns showed no significant differences between the groups (HDL-C, P=0.230; alcohol consumption, P=0.432). The prevalence of hypertension, diabetes mellitus, and dyslipidemia was significantly higher in postmenopausal women (P<0.001).
Association between menopausal status and SLD
Logistic regression analyses assessing the association between menopausal status and probability of SLD are summarized in Table 2. In the unadjusted analysis, menopause was significantly associated with a higher probability of SLD (OR, 1.22; 95% CI, 1.108–1.335; P<0.001). After adjusting for age, BMI, waist circumference, smoking status, alcohol consumption and physical activity, this association remained statistically significant (adjusted model 1: OR, 1.29; 95% CI, 1.079–1.537; P=0.005). In model 2, which additionally included hypertension, diabetes mellitus, and dyslipidemia as covariates, menopause remained significantly associated with an increased probability of SLD (OR, 1.22; 95% CI, 1.014–1.474; P=0.035). When adjusting for menopausal duration, menopausal status remained significantly associated with a higher probability of SLD (OR, 1.40; 95% CI, 1.104–1.764; P=0.005), while menopausal duration itself was not independently associated with SLD probability. This suggests that the observed association was not driven by the duration since menopause.
Subgroup analysis
Subgroup analyses were conducted to evaluate potential effect of modifying the association between menopausal status and SLD probability (Table 3). A statistically significant modifying effect of age was observed (P=0.001), indicating that the strength of the association between menopause and probability of SLD varied across age groups. As illustrated in Figure 2, postmenopausal women exhibited a higher predicted probability of SLD than premenopausal women, particularly in the younger age range of 40 to 49 years. This difference gradually diminished with advancing age, with convergence of the predicted probability curves observed at approximately 57 to 59 years of age.
No significant effect modification was observed for other factors, including BMI, waist circumference, smoking status, physical activity, hypertension, diabetes mellitus, or dyslipidemia, indicating that the association between menopausal status and the probability of SLD was typically consistent across these subgroups.
When BMI and waist circumference were analyzed as categorical variables using clinically relevant cutoffs (BMI ≥25 kg/m2 and waist circumference ≥85 cm), the overall pattern of age-related effect modification in the association between menopausal status and SLD probability remained consistent with the primary analysis (Supplement 1).
Discussion
In this large, nationally representative sample of Korean women aged 40 to 59 years, we found that menopausal status was independently associated with a higher probability of SLD, even after adjusting for age, obesity, and traditional metabolic risk factors. The probability-based prevalence of SLD was significantly higher in postmenopausal women compared with premenopausal women; and multivariate logistic regression analyses consistently demonstrated that menopausal status was independently associated with the probability of SLD. Additional adjustment for menopausal duration did not materially alter the association between menopausal status and the probability of SLD, suggesting that the observed metabolic vulnerability may be significantly related to the menopausal transition itself rather than to the cumulative time since menopause. Notably, the strength of the association between menopausal status and the probability of SLD differed across age groups (P for interaction=0.001) and was most pronounced among younger middle-aged women aged 40 to 49 years.
Our findings are consistent with the accumulating evidence that menopause is accompanied by substantial metabolic alterations that increase women’s susceptibility to cardiometabolic diseases. Davis et al. [11] reported that estrogen decline during menopause promotes the visceral adipose tissue preferential accumulation, even without significant weight gain. Previous studies have established connections between reduced estrogen levels and obesity, type 2 diabetes mellitus, and cardiovascular diseases [12,20]. The combination of increased adipose tissue accumulation and estrogen deficiency triggers metabolic changes, including impaired insulin sensitivity and altered lipid metabolism, which increase metabolic disorders susceptibility among postmenopausal women [21,22]. Furthermore, menopausal transition is associated with increased systemic inflammation, marked by elevated levels of C-reactive protein, interleukin-6, and tumor necrosis factor-alpha. This proinflammatory state contributes to metabolic dysfunction and creates a systemic environment that promotes metabolic liver disease development [23,24].
Beyond these systemic metabolic changes, menopause may influence the probability of SLD through liver-specific mechanisms mediated by estrogen signaling. Estrogen exerts protective effects on the liver by promoting fatty acid β-oxidation and suppressing de novo lipogenesis via receptor-mediated pathways. Estrogen deficiency has been shown to impair lipid oxidation, upregulates lipogenic gene expression (such as SREBP-1c, FASN, and ACC), and alters adipokine secretion, thereby promoting hepatic steatosis in postmenopausal women [25]. Collectively, these hepatic mechanisms provide biological plausibility for the observed association between menopausal status and increased probability of SLD.
Previous studies have reported NAFLD higher prevalence in postmenopausal compared with premenopausal women; and estrogen deficiency has been implicated in the acceleration of hepatic steatosis [26]. Ryu et al. [13] also demonstrated that NAFLD prevalence progressively increased across menopausal stages after adjusting for age, BMI, and metabolic factors. Our study expands upon these findings by applying the updated definition of SLD, which explicitly incorporated metabolic dysfunction into the disease classification, and by analyzing a larger, nationally representative sample of Korean women. In contrast to previous studies focusing primarily on the menopausal transition stages, our study directly compared premenopausal and postmenopausal women, thereby providing more robust population-level evidence that menopausal status is independently associated with higher probability of metabolically driven SLD under the revised nomenclature.
An important finding of this study was that the association between menopausal status and SLD probability was modified by age. The strongest association was observed among women in their 40s, with a gradual attenuation in later age groups. This age-dependent pattern suggests that menopausal transition confers an additional metabolic burden that increases the probability of SLD beyond the effects of chronological aging alone, particularly during early midlife. As age advances, factors such as sarcopenia, reduced physical activity, and cumulative metabolic deterioration may increasingly influence the probability of SLD, thereby attenuating the relative contribution of menopausal status. Importantly, while this effect modification does not contradict the independent association observed in multivariate models, it provides an important biological and clinical context. Clinically, these findings indicate that women undergoing early midlife menopause represent a particularly vulnerable subgroup. This underscores the importance of early metabolic screening and targeted preventive strategies during the initial menopausal transition period.
This study has several strengths. First, it utilized a large, nationally representative sample of 15,106 women from the KNHANES over a 17-year period (2007–2023), employing robust sampling methods that enhance the generalizability of our findings to middle-aged Korean women. Second, comprehensive adjustment for demographic, lifestyle, and metabolic factors strengthened the robustness of the observed associations. Third, we applied the updated SLD nomenclature to better reflect the metabolic nature of the disease, moving beyond the traditional NAFLD definitions to provide a more clinically relevant assessment. Finally, the identification of age-related effect modifications refines risk stratification and has potential implications for targeted prevention strategies.
However, this study has several limitations. First, the cross-sectional design precludes the inference of causality and assessment of SLD progression or the impact of menopausal duration on disease severity. Longitudinal studies are needed to confirm the temporal relationships between menopause and SLD. Second, SLD was assessed using a validated surrogate marker rather than imaging or histological confirmation; therefore, our findings reflect probability-based risk rather than a definitive diagnosis. Third, menopausal status was assessed based on self-reporting without hormonal confirmation, which may have introduced recall bias. Additionally, information on hysterectomy and hormone therapy was not available for all survey cycles, although the proportion of affected participants was relatively small to the overall sample. Finally, residual confounding by unmeasured factors, such as dietary patterns, stress levels, and genetic predisposition, cannot be excluded.
In conclusion, our study demonstrated that menopausal status was independently associated with higher probability of SLD among middle-aged Korean women, an association most pronounced during early midlife. These findings highlight the menopausal transition as a clinically meaningful period for metabolic risk stratification rather than for disease diagnosis. While postmenopausal women tend to exhibit higher SLD risk profiles, our results suggest that menopause occurring in the 40s may serve as an early warning signal of heightened metabolic vulnerability. From clinical and public health perspectives, incorporating menopausal status into targeted screening and preventive strategies may facilitate timely lifestyle and metabolic interventions. Therefore, this approach could help mitigate the long-term burden of metabolically driven liver disease in the expanding postmenopausal population.
Notes
Conflict of interest
Su Hwan Cho serves as an Associate Editor of the Korean Journal of Family Medicine but has no role in the decision to publish this article. Except for that, no other potential conflict of interest relevant to this article was reported.
Acknowledgments
The authors used a generative artificial intelligence (AI) tool exclusively for English language editing to enhance clarity during manuscript preparation. The AI tool was not used for study design, data analysis, data interpretation, or development of scientific content. All interpretations and conclusions were determined by the authors, who critically reviewed and finalized the manuscript. The authors take full responsibility for its integrity and accuracy.
Funding
None.
Data availability
Data of this research are available from the corresponding author upon reasonable request.
Author contribution
Conceptualization: all authors. Methodology: all authors. Software: HSK. Validation: SHC. Formal analysis: HSK. Investigation: SHC. Resources: all authors. Data curation: HSK. Project administration: SHC. Visualization: SHC. Supervision: SHC. Writing– original draft: HSK. Writing–review & editing: all authors. Final approval of the manuscript: all authors.
Flowchart of study population selection. Flow diagram of participant selection from the Korea National Health and Nutrition Examination Survey (KNHANES), 2007–2023. Of 21,653 women aged 40–59 years, those with incomplete data (n=5,556), fasting time ≤8 hours (n=364), history of hysterectomy (n=301), current use of hormone therapy (n=311), or other causes of steatotic liver disease (n=15) were excluded. The final analytic sample included 15,106 women.
Figure. 2.
Predicted probability of steatotic liver disease (SLD) by age and menopausal status. Predicted probability of SLD across age groups (40–59 years) by menopausal status among Korean women, based on weighted multivariable logistic regression analysis adjusted for age, body mass index, waist circumference, smoking, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia. The difference in the predicted probability between premenopausal and postmenopausal women is greatest among those in their 40s and progressively attenuated with advancing age, with curves converging around the late 50s. P for interaction between menopause and age=0.001. Statistical method: weighted logistic regression with the interaction term for menopause×age adjusted for the covariates listed above.
Values are presented as mean±standard deviation or number (%). The variables used were based on survey results or data provided by KNHANES. A chi-square test and independent t-tests were utilized for comparing general characteristics.
a)Regular exercisers were defined as individuals who engaged in at least 150 min/wk of moderate-intensity aerobic physical activity, or 75 min/wk of vigorous-intensity aerobic physical activity, or an equivalent combination of both.
b)Alcohol consumption was assessed based on the frequency and amount of alcohol intake. Participants were categorized into three groups: non-drinkers or light drinkers (<14 drinks/wk), moderate drinkers (14–35 drinks/wk), and heavy drinkers (>35 drinks/wk).
c)Hypertension was defined as taking antihypertensive medication or having a blood pressure of 140/90 mm Hg or higher on the day of examination. Diabetes mellitus was defined as taking diabetes medications or having a fasting blood glucose of 126 mg/dL or higher or an HbA1c of 6.5 or higher. Dyslipidemia was defined as taking medication or having a low-density cholesterol level of 130 mg/dL or higher.
Table 2.
Association between steatotic liver disease and menopausal status
Weighted multivariable logistic regression test was employed to assess the correlation between steatotic liver disease and menopausal status. Physical activity was defined at least 150 min/wk of moderate-intensity, or 75 min/wk of vigorous-intensity aerobic physical activity, or an equivalent combination of moderate- and vigorous-intensity aerobic activity. Smoking status was categorized as never, former, or current smoker. Alcohol consumption was assessed based on the frequency and amount of alcohol intake. Participants were categorized into three groups: non-drinkers or light drinkers (<14 drinks/wk), moderate drinkers (14–35 drinks/wk), and heavy drinkers (>35 drinks/wk).
OR, odds ratio; CI, confidence interval; BMI, body mass index.
a)Adjusted for age, BMI, waist circumference, smoking status, alcohol consumption, and physical activity.
b)Adjusted for age, BMI, waist circumference, smoking status, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia.
Table 3.
Subgroup analysis of the association between steatotic liver disease and menopausal status
Subgroup
OR (95% CI)
P for interaction
Age
0.933 (0.896–0.971)
0.001
Body mass index
1.001 (0.925–1.084)
0.971
Waist circumference
1.000 (0.974–1.026)
0.989
Alcohol consumption
Non- or light drinker
1.244 (1.028–1.507)
Moderate drinker
0.676 (0.175–2.606)
0.703
Heavy drinker
1.256 (0.250–6.322)
0.881
Smoking status
No
1.273 (1.046–1.549)
Current
0.665 (0.302–1.463)
0.103
Ex
1.677 (0.595–4.730)
0.293
Physical activity
0.141
Inadequate
1.423 (1.118–1.813)
Adequate
0.986 (0.733–1.327)
Hypertension
0.232
No
1.314 (1.040–1.659)
Yes
1.054 (0.777–1.431)
Diabetes mellitus
0.125
No
1.268 (1.029–1.562)
Yes
1.079 (0.720–1.615)
Dyslipidemia
0.109
No
1.421 (1.045–1.932)
Yes
1.120 (0.887–1.415)
Weighted multivariable logistic regression analyses were performed to evaluate the interaction between menopausal status and each subgroup variable using Korea National Health and Nutrition Examination Survey sampling weights to account for the complex survey design. The model was adjusted for age, body mass index, waist circumference, smoking status, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia unless the variable was used as a stratification factor. P for interaction was obtained from the cross-product term between menopause and each subgroup variable in the regression model.
OR, odds ratio; CI, confidence interval.
References
1. Ho GJ, Tan FX, Sasikumar NA, Tham EK, Ko D, Kim DH, et al. High global prevalence of steatotic liver disease and associated subtypes: a meta-analysis. Clin Gastroenterol Hepatol 2025;23:2423-32.
2. Hardy T, Wonders K, Younes R, Aithal GP, Aller R, Allison M, et al. The European NAFLD Registry: a real-world longitudinal cohort study of nonalcoholic fatty liver disease. Contemp Clin Trials 2020;98:106175.
4. Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global epidemiology of nonalcoholic fatty liver disease: meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology 2016;64:73-84.
6. Eslam M, Newsome PN, Sarin SK, Anstee QM, Targher G, Romero-Gomez M, et al. A new definition for metabolic dysfunction-associated fatty liver disease: an international expert consensus statement. J Hepatol 2020;73:202-9.
8. Im HJ, Ahn YC, Wang JH, Lee MM, Son CG. Systematic review on the prevalence of nonalcoholic fatty liver disease in South Korea. Clin Res Hepatol Gastroenterol 2021;45:101526.
10. Liu X, Chang Y, Jia F, Li Y, Wang Y, Cui J. Association of Life’s Essential 8 with metabolic dysfunction-associated steatotic liver disease (MASLD), a cross-sectional study from the NHANES 2003-2018. Sci Rep 2024;14:17188.
14. Jinjuvadia R, Antaki F, Lohia P, Liangpunsakul S. The association between nonalcoholic fatty liver disease and metabolic abnormalities in the United States population. J Clin Gastroenterol 2017;51:160-6.
16. Lee JH, Kim D, Kim HJ, Lee CH, Yang JI, Kim W, et al. Hepatic steatosis index: a simple screening tool reflecting nonalcoholic fatty liver disease. Dig Liver Dis 2010;42:503-8.
17. Haam JH, Kim BT, Kim EM, Kwon H, Kang JH, Park JH, et al. Diagnosis of obesity: 2022 update of clinical practice guidelines for obesity by the Korean Society for the Study of Obesity. J Obes Metab Syndr 2023;32:121-9.
19. Kweon S, Kim Y, Jang MJ, Kim Y, Kim K, Choi S, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol 2014;43:69-77.
20. Mauvais-Jarvis F. Estrogen and androgen receptors: regulators of fuel homeostasis and emerging targets for diabetes and obesity. Trends Endocrinol Metab 2011;22:24-33.
22. Chedraui P, Escobar GS, Perez-Lopez FR, Palla G, Montt-Guevara M, Cecchi E, et al. Angiogenesis, inflammation and endothelial function in postmenopausal women screened for the metabolic syndrome. Maturitas 2014;77:370-4.
23. Sites CK, Toth MJ, Cushman M, L'Hommedieu GD, Tchernof A, Tracy RP, et al. Menopause-related differences in inflammation markers and their relationship to body fat distribution and insulin-stimulated glucose disposal. Fertil Steril 2002;77:128-35.
24. McCarthy M, Raval AP. The peri-menopause in a woman’s life: a systemic inflammatory phase that enables later neurodegenerative disease. J Neuroinflammation 2020;17:317.
26. Klair JS, Yang JD, Abdelmalek MF, Guy CD, Gill RM, Yates K, et al. A longer duration of estrogen deficiency increases fibrosis risk among postmenopausal women with nonalcoholic fatty liver disease. Hepatology 2016;64:85-91.
Association between menopausal status and the probability of steatotic liver disease in middle-aged Korean women: a cross-sectional study using the Korea National Health and Nutrition Examination Survey 2007–2023
Figure. 1. Flowchart of study population selection. Flow diagram of participant selection from the Korea National Health and Nutrition Examination Survey (KNHANES), 2007–2023. Of 21,653 women aged 40–59 years, those with incomplete data (n=5,556), fasting time ≤8 hours (n=364), history of hysterectomy (n=301), current use of hormone therapy (n=311), or other causes of steatotic liver disease (n=15) were excluded. The final analytic sample included 15,106 women.
Figure. 2. Predicted probability of steatotic liver disease (SLD) by age and menopausal status. Predicted probability of SLD across age groups (40–59 years) by menopausal status among Korean women, based on weighted multivariable logistic regression analysis adjusted for age, body mass index, waist circumference, smoking, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia. The difference in the predicted probability between premenopausal and postmenopausal women is greatest among those in their 40s and progressively attenuated with advancing age, with curves converging around the late 50s. P for interaction between menopause and age=0.001. Statistical method: weighted logistic regression with the interaction term for menopause×age adjusted for the covariates listed above.
Graphical abstract
Figure. 1.
Figure. 2.
Graphical abstract
Association between menopausal status and the probability of steatotic liver disease in middle-aged Korean women: a cross-sectional study using the Korea National Health and Nutrition Examination Survey 2007–2023
Characteristic
Women (n=15,106)
P-value
Premenopause (n=10,346)
Postmenopause (n=4,760)
Age (y)
47.48±5.20
54.67±3.36
<0.001
Body mass index (kg/m2)
23.56±3.57
23.78±3.39
<0.001
Waist circumference (cm)
78.61±9.29
80.33±9.02
<0.001
Systolic blood pressure (mm Hg)
113.85±15.09
118.25±16.56
<0.001
Diastolic blood pressure (mm Hg)
74.87±9.63
76.20±9.55
<0.001
Fasting glucose (mg/dL)
97.52±21.52
101.41±25.07
<0.001
HbA1c (%)
5.65±0.77
5.81±0.83
<0.001
Total cholesterol (mg/dL)
196.40±35.52
205.70±39.20
<0.001
Triglycerides (mg/dL)
113.14±81.09
124.19±92.08
<0.001
HDL-C (mg/dL)
56.01±13.35
56.29±13.81
0.230
LDL-C (mg/dL)
119.60±31.66
126.58±35.47
<0.001
ALT (IU/L)
18.24±16.41
21.48±18.90
<0.001
AST (IU/L)
20.69±12.36
23.56±14.68
<0.001
Regular exercisera)
4,064 (39.28)
2,092 (43.95)
<0.001
Smoking status
<0.001
Non-smoker
9,243 (89.34)
4,347 (91.32)
Current smoker
514 (4.97)
208 (4.37)
Ex-smoker
589 (5.69)
205 (4.31)
Alcohol consumptionb)
0.432
Non- or light drinker
9,928 (95.96)
4,582 (96.26)
Moderate drinker
323 (3.12)
131 (2.75)
Heavy drinker
95 (0.92)
47 (0.99)
Comorbidityc)
Hypertension
1,774 (17.15)
1,354 (28.45)
<0.001
Diabetes mellitus
806 (7.79)
619 (13.00)
<0.001
Dyslipidemia
4,127 (39.89)
2,835 (59.56)
<0.001
Steatotic liver disease
2,151 (20.79)
1,188 (24.96)
<0.001
Model
OR (95% CI)
P-value
Unadjusted
1.22 (1.108–1.335)
<0.001
Adjusted model 1a)
1.29 (1.079–1.537)
0.005
Adjusted model 2b)
1.22 (1.014–1.474)
0.035
Subgroup
OR (95% CI)
P for interaction
Age
0.933 (0.896–0.971)
0.001
Body mass index
1.001 (0.925–1.084)
0.971
Waist circumference
1.000 (0.974–1.026)
0.989
Alcohol consumption
Non- or light drinker
1.244 (1.028–1.507)
Moderate drinker
0.676 (0.175–2.606)
0.703
Heavy drinker
1.256 (0.250–6.322)
0.881
Smoking status
No
1.273 (1.046–1.549)
Current
0.665 (0.302–1.463)
0.103
Ex
1.677 (0.595–4.730)
0.293
Physical activity
0.141
Inadequate
1.423 (1.118–1.813)
Adequate
0.986 (0.733–1.327)
Hypertension
0.232
No
1.314 (1.040–1.659)
Yes
1.054 (0.777–1.431)
Diabetes mellitus
0.125
No
1.268 (1.029–1.562)
Yes
1.079 (0.720–1.615)
Dyslipidemia
0.109
No
1.421 (1.045–1.932)
Yes
1.120 (0.887–1.415)
Table 1. Baseline characteristics of the study population
Values are presented as mean±standard deviation or number (%). The variables used were based on survey results or data provided by KNHANES. A chi-square test and independent t-tests were utilized for comparing general characteristics.
Regular exercisers were defined as individuals who engaged in at least 150 min/wk of moderate-intensity aerobic physical activity, or 75 min/wk of vigorous-intensity aerobic physical activity, or an equivalent combination of both.
Alcohol consumption was assessed based on the frequency and amount of alcohol intake. Participants were categorized into three groups: non-drinkers or light drinkers (<14 drinks/wk), moderate drinkers (14–35 drinks/wk), and heavy drinkers (>35 drinks/wk).
Hypertension was defined as taking antihypertensive medication or having a blood pressure of 140/90 mm Hg or higher on the day of examination. Diabetes mellitus was defined as taking diabetes medications or having a fasting blood glucose of 126 mg/dL or higher or an HbA1c of 6.5 or higher. Dyslipidemia was defined as taking medication or having a low-density cholesterol level of 130 mg/dL or higher.
Table 2. Association between steatotic liver disease and menopausal status
Weighted multivariable logistic regression test was employed to assess the correlation between steatotic liver disease and menopausal status. Physical activity was defined at least 150 min/wk of moderate-intensity, or 75 min/wk of vigorous-intensity aerobic physical activity, or an equivalent combination of moderate- and vigorous-intensity aerobic activity. Smoking status was categorized as never, former, or current smoker. Alcohol consumption was assessed based on the frequency and amount of alcohol intake. Participants were categorized into three groups: non-drinkers or light drinkers (<14 drinks/wk), moderate drinkers (14–35 drinks/wk), and heavy drinkers (>35 drinks/wk).
OR, odds ratio; CI, confidence interval; BMI, body mass index.
Adjusted for age, BMI, waist circumference, smoking status, alcohol consumption, and physical activity.
Adjusted for age, BMI, waist circumference, smoking status, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia.
Table 3. Subgroup analysis of the association between steatotic liver disease and menopausal status
Weighted multivariable logistic regression analyses were performed to evaluate the interaction between menopausal status and each subgroup variable using Korea National Health and Nutrition Examination Survey sampling weights to account for the complex survey design. The model was adjusted for age, body mass index, waist circumference, smoking status, alcohol consumption, physical activity, hypertension, diabetes mellitus, and dyslipidemia unless the variable was used as a stratification factor. P for interaction was obtained from the cross-product term between menopause and each subgroup variable in the regression model.
