Cardiovascular Risk Associated With General and Abdominal Obesity in Young Adults Without Traditional Risk Factors: A Korean Nationwide Cohort Study

Abstract

Objective

Body mass index (BMI) is widely used to classify obesity but does not adequately characterize central adiposity. Waist-to-height ratio (WHtR) is a simple anthropometric marker of central obesity that may detect cardiovascular risk not captured by BMI. This study examined whether WHtR provides prognostic information beyond BMI for cardiovascular risk stratification in young adults.

Methods

We analyzed 1,743,968 adults aged 30–39 years who underwent health screening through the 2009 Korean National Health Insurance Service program. Participants were classified into four groups according to BMI and WHtR: BMI <25 kg/m² and WHtR <0.5; BMI ≥25 kg/m² and WHtR <0.5; BMI <25 kg/m² and WHtR ≥0.5; and BMI ≥25 kg/m² and WHtR ≥0.5. The primary endpoint was a composite of myocardial infarction (MI), stroke, and cardiovascular death. Multivariable Cox proportional hazards models were used to estimate adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs).

Results

During a median follow-up of 14.2 years, 37,120 participants (2.12%) experienced the composite outcome. After multivariable adjustment, participants with normal BMI but elevated WHtR (Group 3) had an increased risk compared with Group 1 (aHR, 1.171; 95% CI, 1.118–1.226), with a slightly higher point estimate than Group 2. A similar pattern was observed for MI, whereas stroke risk was more strongly associated with elevated BMI.

Conclusion

Elevated WHtR was associated with increased risk of MI and the composite outcome among young adults with normal BMI. WHtR may provide information complementary to BMI for early cardiovascular risk stratification.

INTRODUCTION

Body mass index (BMI) has long been used as a standard measure for classifying obesity and estimating cardiovascular risk. However, BMI has important limitations because it cannot distinguish lean mass from fat mass and provides little information about fat distribution. As a result, BMI-based classification may underestimate or overestimate excess adiposity and may fail to identify obesity-related metabolic dysfunction.¹,² To address these limitations, alternative anthropometric indices, including waist circumference (WC), waist-to-hip ratio (WHR), and waist-to-height ratio (WHtR), have been introduced.³,⁴,⁵,⁶,⁷

Among these indices, WHtR is a simple and reproducible measure of central obesity that accounts for both WC and height. Previous studies and meta-analyses have consistently shown that abdominal obesity indices, including WHtR, outperform BMI for identifying individuals with increased cardiometabolic risk.⁴,⁵,⁶ A WHtR cutoff of 0.5 has also been widely used as a practical threshold for central adiposity screening, although the optimal threshold may differ by ethnicity, age, and population characteristics.⁴,⁵,⁸,⁹,¹⁰ The European Atherosclerosis Society recently proposed a pathophysiology-based framework for systemic metabolic disorders that emphasizes visceral and ectopic fat accumulation as key drivers of cardiometabolic risk, even before overt clinical disease develops.¹¹

Previous studies have shown that WHtR is more strongly associated with cardiometabolic risk factors, such as diabetes mellitus, hypertension, and dyslipidemia, than BMI or WC alone.⁵,⁶,¹² More recently, large prospective cohort studies have reported significant associations between WHtR and incident cardiovascular disease.¹³,¹⁴ However, most studies of anthropometric indices and cardiovascular risk have focused on middle-aged populations or individuals with established cardiometabolic risk factors. In young adults, hypertension, diabetes mellitus, and dyslipidemia may not yet be clinically evident despite early visceral adiposity and metabolic dysfunction. Central obesity may precede traditional cardiometabolic disease through insulin resistance, systemic inflammation, and endothelial dysfunction, thereby contributing to early cardiovascular vulnerability even in otherwise low-risk individuals.¹⁵,¹⁶ Evaluating WHtR in young adults without established traditional risk factors may therefore help identify hidden cardiometabolic risk that is not captured by BMI alone.⁵ On this basis, we investigated whether WHtR provides prognostic information beyond BMI in a large nationwide cohort of Korean adults aged 30–39 years who participated in the 2009 National Health Screening Program and were free of traditional cardiovascular risk factors.

MATERIALS AND METHODS

This study was reported in accordance with the REporting of studies Conducted using Observational Routinely collected health Data guideline.

1. Study design and participant selection

This study used data from the Korean National Health Insurance Service (K-NHIS). The database covers approximately 97% of the Korean population and consists predominantly of individuals of East Asian ethnicity. The K-NHIS database contains comprehensive information on demographic characteristics, laboratory results, diagnostic codes, hospitalization records, and medication prescriptions. Analyses using these claims-based data therefore capture health care use and clinical outcomes in routine practice.¹⁷

We included adults aged 30–39 years who participated in the 2009 Korean National Health Screening Program. Individuals were excluded if baseline questionnaire data, anthropometric measurements, or laboratory test results were missing. Hypertension was defined as systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg measured during the health screening examination, or as the presence of International Classification of Diseases, Tenth Revision (ICD-10) codes I10–I13 or I15 accompanied by at least one prescription for antihypertensive medication. Diabetes mellitus was defined as fasting blood glucose ≥126 mg/dL or the presence of ICD-10 codes E11–E14 with at least one prescription for antidiabetic medication. Dyslipidemia was defined as low-density lipoprotein cholesterol (LDL-C) ≥160 mg/dL or the presence of ICD-10 code E78 with a prescription for lipid-lowering medication.¹⁸,¹⁹ To construct a cohort free of traditional cardiometabolic risk factors, we excluded individuals who met any of the above criteria for hypertension, diabetes mellitus, or dyslipidemia before the index health screening. Individuals with a history of myocardial infarction (MI) or stroke before the index screening were also excluded. To minimize reverse causality and reduce potential bias related to subclinical disease at baseline, participants who developed study outcomes within the first year after the index health screening were excluded through a prespecified 1-year lag period. This study was approved by the Institutional Review Board (IRB) of Dongtan Sacred Heart Hospital (IRB number: HDT 2024-02-003). Participants who underwent the national health examination provided written informed consent for the use of their data for research purposes.

2. Definition of covariates

Demographic variables included age, sex, body weight, and height. Clinical variables included systolic and diastolic blood pressure, and laboratory variables included renal function, fasting glucose, total cholesterol, LDL-C, and high-density lipoprotein cholesterol. Renal function was assessed using the estimated glomerular filtration rate (eGFR), which was calculated with the Modification of Diet in Renal Disease equation. Chronic kidney disease was defined as an eGFR <60 mL/min/1.73 m². Lifestyle behaviors, including alcohol consumption, physical activity, and smoking status, were assessed using self-reported questionnaires. Alcohol consumption was defined as any alcohol intake >0 g/day. Regular exercise was defined as moderate-to-vigorous physical activity on ≥4 days per week. Low-income status was defined as coverage by the Medical Aid program for the lowest-income population or membership in the lowest 20% of individuals registered in the National Health Insurance system based on monthly household income.

3. Main exposure

BMI was calculated as body weight in kilograms divided by height in meters squared (kg/m²). WHtR was calculated as WC divided by height, with both measurements expressed in centimeters. Participants were categorized into four groups according to combined BMI and WHtR status: Group 1, BMI <25 kg/m² and WHtR <0.5 (reference group); Group 2, BMI ≥25 kg/m² and WHtR <0.5; Group 3, BMI <25 kg/m² and WHtR ≥0.5; and Group 4, BMI ≥25 kg/m² and WHtR ≥0.5. Fig. 1 presents the study flow and overall design.

Fig. 1. Flowchart of study population selection.

MI, myocardial infarction; HTN, hypertension; DM, diabetes mellitus; BMI, body mass index; WHtR, waist-to-height ratio.

4. Outcome definition

The primary outcome was a composite of stroke, MI, and cardiovascular death. Secondary outcomes included each component of the primary outcome and all-cause mortality. Stroke was defined as hospitalization for more than 4 days with ICD-10 codes I60–I63 and a claim for brain imaging. MI was defined as hospitalization for more than 4 days with ICD-10 codes I21–I22 or claims for relevant diagnostic procedures. Cardiovascular death was identified from cause-of-death codes in the national mortality registry maintained by Statistics Korea, and all-cause mortality was obtained from the same national death records. In Korea, mortality data are derived from residential registration records maintained by Statistics Korea through the Korean Statistical Information Service, which provides official national and local population and mortality data. According to annual reports from Statistics Korea, the accuracy of recorded cause of death is approximately 92%.²⁰ Participants were followed until December 31, 2022, or until development of the primary outcome, whichever occurred first.

5. Statistical analysis

Continuous variables are presented as means ± standard deviations or medians with interquartile ranges, and categorical variables are reported as numbers and percentages. Baseline characteristics were compared using one-way analysis of variance or the Kruskal–Wallis test for continuous variables and the χ² test or Fisher exact test for categorical variables, as appropriate. Multivariable Cox proportional hazards regression analyses were used to estimate adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for the association between the four BMI-WHtR groups and each outcome. Four models were used for multivariable analysis: Model 1 was unadjusted; Model 2 was adjusted for age and sex; Model 3 was adjusted for age, sex, alcohol consumption, regular physical activity, low-income status, systolic blood pressure, diastolic blood pressure, and total cholesterol; and Model 4 was adjusted for the variables in Model 3 plus chronic kidney disease, proteinuria, and fasting glucose. Time-to-event outcomes were analyzed using Kaplan–Meier estimates and compared using the log-rank test. Statistical significance was defined as a two-sided p-value <0.05. Analyses were performed using SAS software, version 9.4 (SAS Institute Inc.).

RESULTS

A total of 2,032,253 participants aged 30–39 years who underwent national health screening in 2009 were initially included. After excluding individuals with missing health screening data (n=70,391), a prior history of stroke or MI (n=5,083), or pre-existing hypertension, diabetes mellitus, or dyslipidemia (n=210,502), and after applying a 1-year washout period (n=2,309), 1,743,968 participants were included in the final analysis. Participants were stratified into four groups according to combined BMI and WHtR status: Group 1, 1,122,443 participants (64.3%); Group 2, 178,352 participants (10.2%); Group 3, 86,391 participants (4.9%); and Group 4, 356,782 participants (20.4%) (Fig. 1).

Baseline demographic and clinical characteristics according to the four BMI-WHtR groups are summarized in Table 1. Overall, 72.3% of the study population was male, and the mean age was 34.4±2.9 years. The high-BMI groups (Groups 2 and 4) showed a marked male predominance, with approximately 90% of participants in these groups being male. Smoking prevalence was also higher in the high-BMI groups, with nearly half of participants reporting current smoking at the time of health screening. Among lipid parameters, Group 4 had the highest total cholesterol level, at 204.28±39.75 mg/dL.

Table 1. Demographic characteristics of the study population according to the BMI-WHtR groups.

Characteristics	Total (n=1,743,968)	Group 1 (n=1,122,443)	Group 2 (n=178,352)	Group 3 (n=86,391)	Group 4 (n=356,782)	p-value
Age (yr)	34.4±2.9	34.29±2.91	34.27±2.87	34.98±2.86	34.64±2.87	<0.0001
Male sex	1,261,512 (72.3)	731,526 (65.2)	157,358 (88.2)	60,095 (69.6)	312,533 (87.6)	<0.0001
Height (cm)	169.2±8.0	168.47±8.15	172.47±7.07	166.28±7.91	170.61±7.33	<0.0001
Weight (kg)	67.7±12.6	61.6±9.12	77.98±7.26	65.55±7.67	81.87±10.18	<0.0001
BMI (kg/m2)	23.5±3.4	21.6±1.98	26.17±1.12	23.62±1.19	28.07±2.5	<0.0001
Current smoker	716,180 (41.1)	418,207 (37.3)	86,938 (48.7)	32,708 (37.9)	178,327 (50.0)	<0.0001
Alcohol consumption	534,422 (30.6)	310,737 (27.7)	64,282 (36.0)	26,756 (31.0)	132,647 (37.2)	<0.0001
Regular exercise	576,265 (33.0)	371,742 (33.1)	63,760 (35.7)	25,228 (29.2)	115,535 (32.4)	<0.0001
Low income	242,129 (13.9)	165,695 (14.8)	18,684 (10.5)	15,361 (17.8)	42,389 (11.9)	<0.0001
Chronic kidney disease	19,499 (1.1)	18,527 (1.7)	182 (0.1)	455 (0.5)	335 (0.1)	<0.0001
Proteinuria	23,911 (1.4)	13,487 (1.2)	2,501 (1.4)	1,141 (1.3)	6,782 (1.9)	<0.0001
SBP (mmHg)	119.4±12.9	116.82±12.19	123.35±12.13	119.33±12.56	125.67±13.09	<0.0001
DBP (mmHg)	75.1±9.3	73.37±8.78	77.51±8.86	75±9.23	79.37±9.66	<0.0001
Glucose (mg/dL)	92.1±16.0	90.45±13.82	93.72±16.61	92.74±17.73	96.08±20.44	<0.0001
Cholesterol (mg/dL)	190.2±37.5	184.12±35.15	197.47±37.85	195.68±37.28	204.28±39.75	<0.0001
LDL-C (mg/dL)	116.8±210.4	113.18±217.01	122.28±210.14	118.09±162.41	124.96±199.19	<0.0001
HDL-C (mg/dL)	55.6±27.7	58.06±28.01	51.7±27.55	54.9±29.81	50.1±25.3	<0.0001
eGFR (mL/min/1.73 m2)	101.5±26.5	93.62±22.24	113.59±24.42	100.89±24.62	120.35±28.45	<0.0001

Values are presented as number (%) or mean ± standard deviation.

BMI, body mass index; WHtR, waist-to-height ratio; SBP, systolic blood pressure; DBP, diastolic blood pressure; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate.

During a median follow-up of 14.21 years, the cumulative incidence of the primary composite outcome was 2.12%. The incidences of stroke, MI, cardiovascular death, and all-cause mortality were 0.8%, 1.29%, 0.14%, and 1.03%, respectively. In unadjusted analyses, the risk estimate for the composite outcome was slightly higher in Group 2 than in Group 3 (Group 2: aHR, 1.387; 95% CI, 1.342–1.434; Group 3: aHR, 1.336; 95% CI, 1.276–1.399). A similar pattern was observed for stroke (Group 2: aHR, 1.601; 95% CI, 1.519–1.687; Group 3: aHR, 1.362; 95% CI, 1.261–1.470). In contrast, the risk estimate for MI was modestly higher in Group 3 than in Group 2 (Group 2: aHR, 1.265; 95% CI, 1.211–1.321; Group 3: aHR, 1.328; 95% CI, 1.253–1.407) (Table 2). These findings are shown in Fig. 2.

Table 2. Multivariable analysis of the association between the BMI-WHtR groups and clinical outcomes.

BMI-WHtR groups		Total number	Clinical outcomes	Duration (PY)	IR per 1,000 PY	HR (95% CI)
						Model 1	Model 2	Model 3	Model 4
Composite outcome
	Group 1	1,122,443	19,564	15,915,741.05	1.23	1 (Ref.)	1 (Ref.)	1 (Ref.)	1 (Ref.)
	Group 2	178,352	4,289	2,519,874.42	1.70	1.387 (1.342–1.434)	1.292 (1.250–1.336)	1.167 (1.128–1.207)	1.160 (1.122–1.200)
	Group 3	86,391	2,003	1,221,158.24	1.64	1.336 (1.276–1.399)	1.250 (1.194–1.309)	1.177 (1.124–1.233)	1.171 (1.118–1.226)
	Group 4	356,782	11,264	5,023,247.50	2.24	1.830 (1.788–1.873)	1.661 (1.622–1.701)	1.396 (1.362–1.430)	1.377 (1.343–1.411)
Stroke
	Group 1	1,122,443	6,951	15,980,276.73	0.43	1 (Ref.)	1 (Ref.)	1 (Ref.)	1 (Ref.)
	Group 2	178,352	1,761	2,532,921.90	0.70	1.601 (1.519–1.687)	1.428 (1.354–1.505)	1.316 (1.248–1.388)	1.307 (1.24–1.379)
	Group 3	86,391	726	1,227,451.01	0.59	1.362 (1.261–1.470)	1.265 (1.172–1.366)	1.21 (1.121–1.307)	1.203 (1.114–1.298)
	Group 4	356,782	4,562	5,056,465.59	0.90	2.08 (2.004–2.159)	1.812 (1.744–1.882)	1.571 (1.51–1.634)	1.549 (1.489–1.612)
Myocardial infarction
	Group 1	1,122,443	12,265	15,946,766.77	0.77	1 (Ref.)	1 (Ref.)	1 (Ref.)	1 (Ref.)
	Group 2	178,352	2,456	2,528,135.58	0.97	1.265 (1.211–1.321)	1.219 (1.166–1.273)	1.091 (1.044–1.141)	1.085 (1.039–1.135)
	Group 3	86,391	1,249	1,224,509.328	1.02	1.328 (1.253–1.407)	1.248 (1.178–1.323)	1.169 (1.103–1.239)	1.163 (1.097–1.233)
	Group 4	356,782	6,514	5,044,522.92	1.29	1.683 (1.633–1.735)	1.577 (1.529–1.626)	1.309 (1.268–1.352)	1.29 (1.249–1.332)
Cardiovascular death
	Group 1	1,122,443	1,181	16,012,541.96	0.07	1 (Ref.)	1 (Ref.)	1 (Ref.)	1 (Ref.)
	Group 2	178,352	275	2,541,535.91	0.11	1.468 (1.288–1.674)	1.246 (1.092–1.422)	1.047 (0.917–1.196)	1.049 (0.918–1.198)
	Group 3	86,391	108	1,230,925.61	0.09	1.191 (0.987–1.45)	1.08 (0.886–1.315)	0.97 (0.796–1.182)	0.972 (0.797–1.184)
	Group 4	356,782	801	5,078,592.74	0.16	2.142 (1.958–2.343)	1.765 (1.611–1.933)	1.293 (1.177–1.422)	1.289 (1.172–1.417)

Model 1: Unadjusted. Model 2: Adjusted for age, sex. Model 3: Adjusted for variables in Model 2 plus alcohol consumption, regular physical activity, low-income level, systolic and diastolic blood pressure and total cholesterol. Model 4: Adjusted for variables in Model 3 plus chronic kidney disease, proteinuria, and fasting glucose.

BMI, body mass index; WHtR, waist-to-height ratio; PY, person-year; IR, incidence rate; HR, hazard ratio; CI, confidence interval.

Fig. 2. Unadjusted Kaplan–Meier curves for the cumulative incidence probability of clinical outcomes according to body mass index-waist-to-height ratio groups. (A) Composite endpoint; (B) Stroke; (C) Myocardial infarction; (D) Cardiovascular death.

After adjustment for all covariates, Group 3 had a slightly higher risk estimate for the primary composite outcome than Group 2, although the estimates were similar (Group 3: aHR, 1.171; 95% CI, 1.118–1.226; Group 2: aHR, 1.160; 95% CI, 1.122–1.200). Group 3 also had a modestly higher risk estimate for MI than Group 2 (Group 3: aHR, 1.163; 95% CI, 1.097–1.233; Group 2: aHR, 1.085; 95% CI, 1.039–1.135). By contrast, Group 2 had a modestly higher risk estimate for stroke than Group 3 (Group 2: aHR, 1.307; 95% CI, 1.240–1.379; Group 3: aHR, 1.203; 95% CI, 1.114–1.298). For cardiovascular death, only Group 4 had a statistically significant increase in risk (aHR, 1.289; 95% CI, 1.172–1.417) (Table 2). All-cause mortality was lower in Groups 2–4 than in Group 1 after full adjustment (Supplementary Table 1, Supplementary Fig. 1).

Sex-stratified analyses showed generally consistent associations between BMI-WHtR groups and cardiovascular outcomes in both men and women (Supplementary Table 2). However, the associations of BMI-WHtR groups with the primary composite outcome and MI were more pronounced in women than in men (p for interaction=0.0003 and 0.031, respectively), whereas the sex interactions for stroke and cardiovascular death were not statistically significant.

DISCUSSION

In this nationwide cohort study, central obesity identified by WHtR was associated with increased long-term cardiovascular risk among young adults without overt traditional cardiometabolic risk factors. Participants with normal BMI but elevated WHtR (Group 3) had modestly higher risk estimates for MI and the primary composite outcome than those with elevated BMI but normal WHtR (Group 2). In contrast, stroke risk appeared more closely related to elevated BMI than to elevated WHtR. Cardiovascular mortality was significantly increased only among participants with both elevated BMI and elevated WHtR.

Many previous studies have suggested that WHtR is more strongly associated with cardiometabolic risk than BMI.⁴,⁵,⁶,⁷,¹²,¹³,¹⁴ In the present study, however, WHtR was not uniformly more strongly associated with cardiovascular outcomes in the unadjusted analysis. This apparent discrepancy may reflect differences in study design and population characteristics rather than the clinical relevance of WHtR itself. Several meta-analyses have shown that WHtR is closely associated with cardiometabolic risk factors and may perform better than BMI and WC.⁴,⁵ Unlike WC alone, WHtR incorporates height into the assessment of abdominal adiposity and therefore evaluates fat distribution relative to body size. These findings suggest that the prognostic value of anthropometric indices may vary according to population characteristics, baseline risk profiles, and specific cardiovascular outcomes. Such variability may be particularly important in younger populations with low baseline cardiometabolic risk.

Participants in previous studies were predominantly middle-aged, with mean ages of approximately 53–56 years.¹³,¹⁴ In young adults, overt hypertension, diabetes mellitus, and dyslipidemia may not yet be clinically apparent despite early visceral adiposity, insulin resistance, and subclinical metabolic dysfunction. Recent pathophysiology-based frameworks emphasize that visceral and ectopic fat accumulation may contribute to insulin resistance, systemic inflammation, and early cardiovascular vulnerability before conventional cardiometabolic disease becomes clinically manifest.¹¹ Similarly, mechanistic data suggest that visceral adiposity is associated with adverse metabolic profiles linked to future type 2 diabetes and MI independently of BMI, supporting the concept that hidden adiposity-related risk may precede overt clinical disease.¹⁵ Reliance on BMI alone may therefore underestimate early cardiovascular vulnerability in otherwise low-risk young adults. In this context, WHtR may better reflect hidden central adiposity and identify individuals with early cardiometabolic risk before traditional risk factors become clinically apparent. Accordingly, the present study focused on adults aged 30–39 years who were free of major cardiometabolic diseases at baseline.

A previous study of a relatively young population with a mean age of approximately 37 years reported that WHtR had greater discriminatory ability than BMI or WC for identifying metabolic risk factors, including diabetes mellitus, hypertension, dyslipidemia, and metabolic syndrome.¹² However, that study evaluated cross-sectional associations with cardiometabolic abnormalities rather than incident cardiovascular outcomes. Extending these prior findings, the present study assessed the association of four BMI-WHtR categories with long-term cardiovascular events. After full adjustment, participants with normal BMI but elevated WHtR had modestly higher risk estimates for MI and the primary composite outcome than those with elevated BMI but normal WHtR. These findings suggest that WHtR may help identify individuals with hidden central obesity and early cardiometabolic vulnerability who would not be classified as obese on the basis of BMI alone.

Sex-stratified analyses showed broadly consistent associations across sexes, although some associations, particularly those for the primary composite outcome and MI, were modestly stronger in women. These differences may partly reflect known sex-related differences in adipose tissue distribution and metabolic responses, which could influence the cardiovascular consequences of central adiposity.²¹

A previous meta-analysis demonstrated that WC, WHR, and WHtR were positively associated with incident stroke, particularly ischemic stroke.²² In a Chinese prospective cohort, higher BMI was associated with increased stroke risk, particularly among younger individuals, although the associations were influenced by coexisting cardiometabolic risks.²³ Similarly, a large nationwide Korean cohort study reported a graded increase in MI and ischemic stroke risk according to WC.²⁴ However, those studies included mostly middle-aged or general adult populations with a higher baseline risk burden. The present study examined adults aged 30–39 years who were free of hypertension, diabetes mellitus, and dyslipidemia at baseline and therefore represented a relatively low-risk population. Stroke in young adults can arise from diverse mechanisms, and not all cases are directly driven by metabolic or atherosclerotic pathways. A substantial proportion of stroke events in young adults are related to nonatherosclerotic mechanisms, such as cardioembolism, arterial dissection, or cryptogenic etiologies.²⁵ In the present study, individuals with baseline hypertension, diabetes mellitus, or dyslipidemia were excluded, and multivariable adjustment further accounted for major vascular risk factors. Under these conditions, the residual independent contribution of central adiposity to incident stroke may have been attenuated.

Large-scale pooled studies have shown that the relationship between adiposity and mortality is complex and may be influenced by competing risks, reverse causality, and baseline population characteristics.²⁶,²⁷,²⁸ Recent systematic reviews have also shown that elevated WHtR is associated not only with cardiometabolic abnormalities and incident cardiovascular disease but also with increased cardiovascular mortality risk.²⁹ In the present study, despite the relatively low absolute number of cardiovascular deaths expected in this young and low-risk population, participants with concomitantly elevated BMI and WHtR had significantly increased cardiovascular mortality. These findings suggest that combined or more advanced adiposity phenotypes may confer measurable long-term cardiovascular mortality risk even in early adulthood.

This study has several limitations. First, although the K-NHIS database provides a large nationwide administrative dataset, the study population was limited to Korean adults aged 30–39 years who participated in the 2009 National Health Screening Program. Therefore, the findings may not fully represent all Korean individuals in their 30s and should be interpreted with caution. Second, this observational study was based on administrative health data; therefore, causal relationships cannot be established, and residual confounding by unmeasured variables, such as diet, physical activity, socioeconomic factors, or cardiorespiratory fitness, cannot be excluded. Third, anthropometric measures were obtained at a single baseline health examination, and changes in body weight or fat distribution over time were not captured, which may have resulted in misclassification during follow-up. Although WHtR ≥0.5 has been widely used in diverse populations, including Asian cohorts, as a practical screening threshold for central adiposity, the optimal cutoff for young Korean adults may differ according to ethnicity-, age-, and population-specific body composition patterns. Alternative population-specific WHtR thresholds may therefore alter group classification and effect estimates, and future studies are warranted to determine more precise thresholds for cardiovascular risk stratification in young Korean adults. Fourth, the study population comprised relatively young adults without baseline hypertension, diabetes mellitus, or dyslipidemia. Although this design allowed evaluation of early cardiovascular risk in a low-risk cohort, the relatively low absolute event rates, particularly for stroke and cardiovascular death, may have reduced the ability to detect modest associations. In addition, because individuals with established cardiometabolic disease were excluded and major vascular risk factors were adjusted for, the observed associations may underestimate the total impact of adiposity on downstream vascular outcomes. Finally, BMI and WHtR are indirect anthropometric indices and do not directly quantify visceral or ectopic fat; imaging-based assessments were not available in the present database. Future studies incorporating repeated measurements of adiposity and metabolic factors may help clarify how early-life obesity influences long-term cardiovascular risk.

In this nationwide cohort of young adults without traditional cardiometabolic risk factors, elevated WHtR was associated with an increased risk of MI and the primary composite outcome even among individuals with normal BMI. These findings suggest that reliance on BMI alone may underestimate cardiovascular risk in young adults and that WHtR may help identify individuals with hidden central obesity who remain at elevated cardiovascular risk despite having normal BMI. Although BMI showed a stronger association with stroke risk, WHtR provided complementary information for identifying early cardiovascular risk, particularly for MI. Incorporating both indices into routine health screening may improve the early identification of at-risk individuals and support more targeted preventive strategies in this otherwise low-risk population.

SUPPLEMENTARY MATERIALS

Supplementary Table 1

Adjusted multivariate analysis about the association between BMI-WHtR groups and all-cause mortality

Supplementary Table 2

Sex-stratified multivariable analyses of the association between BMI-WHtR groups and cardiovascular outcomes

Supplementary Fig. 1

The Kaplan–Meier curves for the cumulative incidence probability of all-cause mortality according to body mass index-waist-to-height ratio groups.