High Waist-to-Height Ratio Increases the Risk of Cardiovascular Outcomes in Adults with Type 1 Diabetes Mellitus: A Nationwide Cohort Study
Article information
Abstract
Background
Central obesity contributes to an increased risk of cardiovascular disease (CVD) and mortality. The waist-to-height ratio (WHtR) is a practical marker of central obesity across sexes, ages, and ethnicities. However, its association with comprehensive cardiovascular (CV) outcomes in patients with type 1 diabetes mellitus (T1DM) remains unclear.
Methods
From a nationwide cohort database (2006–2020), 16,928 Korean adults with T1DM were included. Participants were categorized by their WHtR values using three criteria: a three-group classification (<0.5, 0.5 to <0.6, and ≥0.6) and two binary classifications (≥0.5 vs. <0.5; ≥0.6 vs. <0.6). The primary outcomes were composite CV events, including heart failure (HF), myocardial infarction (MI), ischemic stroke, and CVD-related deaths, with each component analyzed as a secondary outcome.
Results
During a median follow-up of 6.7 years (interquartile range, 5.2 to 8.8), 4,293 composite CV events occurred. Compared to the WHtR <0.5 group, the adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for the composite CV outcome were 1.14 (1.05 to 1.24) in the WHtR 0.5 to <0.6 group and 1.62 (1.38 to 1.90) in the WHtR ≥0.6 group (P for trend <0.001). Increasing trends in aHRs were noted with rising WHtR values for each component of the composite outcome. Compared to the WHtR <0.6 group, the aHRs for the WHtR ≥0.6 group were as follows: HF, 1.49 (95% CI, 1.28 to 1.73); MI, 1.31 (95% CI, 1.02 to 1.68); ischemic stroke, 1.24 (95% CI, 1.02 to 1.51); and CVD-related death, 2.09 (95% CI, 1.49 to 2.92).
Conclusion
High WHtR is associated with an increased risk of CV events in adults with T1DM.
Highlights
• WHtR is associated with CV risk in adults with T1DM.
• CV risk rises stepwise across WHtR categories: <0.5, 0.5–<0.6, and ≥0.6.
• Higher WHtR is linked to increased risk of HF, MI, stroke, and CVD-related death.
• This association was consistent in both men and women.
• WHtR can differentiate CV risk even in those with high BMI or waist circumference.
INTRODUCTION
Central obesity (CO) is associated with insulin resistance, dyslipidemia, high blood pressure, and atherosclerosis [1]. These are the key features of metabolic syndrome (MetS) and are commonly observed in type 2 diabetes mellitus (T2DM) [2-6]. However, studies have confirmed that these factors also contribute to elevated cardiovascular (CV) risk and mortality in type 1 diabetes mellitus (T1DM), demonstrating that T1DM individuals with coexisting MetS face a higher CV risk than those without MetS [7-10]. Considering the increasing prevalence of obesity among patients with T1DM in recent decades [11], it is essential to identify effective CO markers to predict CV outcomes in this population.
The waist-to-height ratio (WHtR) is a practical measure of central adiposity and is recommended by the National Institute for Health and Care Excellence (NICE) [12,13]. Prior research has confirmed that the WHtR is a better estimator of visceral fat than the body mass index (BMI) in patients with T1DM, independent of sex [14]. The WHtR has been suggested as a better screening tool for cardiometabolic risk factors than waist circumference (WC) and BMI in the general population of different nationalities [15]. Compared with WC, which is another index of abdominal obesity, WHtR offers the practical advantages of its simple boundary value in predicting cardiometabolic outcomes and diabetes [16]. The WC is generally larger in men and adults than in women and children, respectively. Conversely, the same boundary values for WHtR can be employed across all sexes and age groups because the WHtR is adjusted for height [16]. Moreover, this feature is globally consistent and covers various ethnicities [16,17].
Despite its importance, the relationship between WHtR and CV outcomes has rarely been investigated in the T1DM population. A European cohort study including 4,668 individuals with T1DM revealed that a WHtR ≥0.5 is associated with an increased risk of heart failure (HF) hospitalization or death [18]. However, to the best of our knowledge, no study has explored the association between the WHtR and composite outcomes that encompass overall CV risks. Thus, we aimed to analyze the impact on a broad range of CV outcomes in a larger population through a nationwide longitudinal study. Moreover, we sought to evaluate the association between changes in the WHtR and the risks of cardiovascular disease (CVD), considering that the WHtR is a modifiable characteristic.
METHODS
Data source and study settings
This study used the Korean National Health Insurance System (KNHIS) database, which is a nationwide cohort covering approximately 97% of Korean citizens [19-25]. It offers demographic and comprehensive medical information, including health checkups with anthropometric and laboratory measurements; healthcare utilization data classified by the International Classification of Disease, 10th Revision (ICD-10); and lifestyle questionnaires covering smoking history, alcohol consumption, and physical activity. The KNHIS data from 2006 to 2020 were employed. We used information from nationwide death certificates provided by the Korean National Statistical Office, classified using ICD-10 codes, to determine the cause of death.
This study was approved by the Institutional Review Board of the Samsung Medical Center (approval no. SMC 2025-02-037) and was conducted in accordance with the Declaration of Helsinki. The requirement for informed consent was waived because the data were publicly accessible and de-identified.
Definition of T1DM and selection of study population
We selected adults (aged ≥19 years) with T1DM through the following process (Fig. 1) [26]: we initially selected individuals who underwent a health examination between January 1, 2009 and December 31, 2015 and had a prior diagnosis of E10 (ICD-10 code for T1DM) before the examination (n=164,314). Next, we included only patients who received three or more insulin prescriptions after E10 diagnosis (n=71,819), and at least one of these prescriptions had to occur within 1–2 years after the diagnosis (n=52,268). Among these individuals, those who had no E10 records within 1–2 years after diagnosis but only had E11–14 (E11, T2DM; E12, malnutrition-related diabetes; E13, other specified diabetes; E14, unspecified diabetes) were excluded, as they were likely to be patients with ketosis-prone T2DM (n=29,599). The remaining 22,669 individuals were considered the T1DM population who underwent a health examination. Among these, individuals with a history of partial or total pancreatectomy before T1DM diagnosis were also excluded (n=37). We further excluded individuals under 20 years of age (n=36), those with a history of CVD before the index date (n=5,658), and those with missing data (n=10). Ultimately, 16,928 adults with T1DM were analyzed.
Inclusion and exclusion flowchart for study participants. ICD-10, International Classification of Disease, 10th Revision; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; CVD, cardiovascular disease.
The index date was set as the date of the health examination between 2009 and 2015; when multiple examinations occurred after T1DM diagnosis, the date closest to the diagnosis was used. All participants were followed-up until the earliest occurrence of any specified CV outcome, death, or end of the cohort period (December 31, 2020).
Anthropometric measurement
In the KNHIS, trained nurses obtained height, weight, and WC values from participants wearing light clothing. The BMI was calculated as weight divided by height squared (kg/m2). The WHtR was calculated as WC/height. According to the NICE guidelines, a WHtR of 0.5 to <0.6 is classified as increased central adiposity (central fat distribution, a range where individuals should ‘consider action’), while a WHtR ≥0.6 is classified as high central adiposity (CO, a range where individuals should ‘take action’) [27-30]. In this study, we performed three separate analyses based on WHtR categories: (1) a three-group comparison using cutoffs of <0.5, 0.5 to <0.6, and ≥0.6; (2) a two-group comparison using 0.5 as the cutoff (≥0.5 vs. <0.5); and (3) another two-group comparison using 0.6 as the cutoff (≥0.6 vs. <0.6).
Two-time point analysis
Two-time point analyses were conducted on individuals who underwent health checkups from 1 to 2 years before the index date (n=5,736). The WHtR values from health checkups before the index date and those measured on the index date were used. WHtR values <0.6 were labeled as CO−, while values ≥0.6 were labeled as CO+. Based on WHtR values from the two-time points, the participants were categorized into four groups and compared: (1) CO− → CO−, (2) CO− → CO+, (3) CO+ → CO−, and (4) CO+ → CO+. Additionally, we conducted analyses assuming WHtR ≥0.5 as the threshold for CO.
Outcomes
Our study’s primary outcome was a composite CV outcome, including incident HF, myocardial infarction (MI), ischemic stroke, and CVD-related deaths. Hospitalization with the following ICD-10 codes was used to define the outcomes: HF (I50), MI (I21–I23), and ischemic stroke (I63). CVD-related deaths were defined as deaths caused by CVD (I00–I99) [31]. Each component of the primary composite outcome was considered a secondary outcome.
Other variables
The KNHIS database provided baseline data on age, sex, weight, BMI, WC, systolic blood pressure (SBP), diastolic blood pressure (DBP), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), aspartate aminotransferase (AST), alanine transaminase (ALT), fasting plasma glucose (FPG), estimated glomerular filtration rate (eGFR), smoking status, alcohol consumption, physical activity, and income level. BMI was categorized using an Asian-specific cutoff point for defining obesity: 25 kg/m2 [32]. Smoking status was categorized as never, past, or current, and alcohol consumption was categorized as none, mild to moderate (<30 g/day for males; <20 g/day for females), or heavy (≥30 g/day for males; ≥20 g/day for females). Physical activity was classified as active if participants reported either engaging in vigorous-intensity exercise for at least 20 minutes per session on 3 or more days per week, or performing moderate-intensity activity for at least 30 minutes per session on 5 or more days per week, based on questionnaire data. Low income was defined as belonging to the bottom 20th percentile. Comorbidities within 1 year of the index date were defined as follows: chronic kidney disease (CKD; eGFR <60 mL/min/1.73 m2), hypertension (HTN; receiving antihypertensive medication or SBP ≥140 mm Hg or DBP ≥90 mm Hg), and dyslipidemia (receiving lipid-modifying agents or TC ≥240 mg/dL). The 2009 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation was used to calculate eGFR for the definition of CKD [33].
Statistical analysis
Categorical variables were reported as percentages, and continuous variables were presented as mean and standard deviation. We compared the baseline characteristics between the WHtR groups using the chi-square or Fisher’s exact test for categorical variables and the analysis of variance for continuous variables.
Kaplan-Meier curves were generated to compare the incidence of CV outcomes stratified by WHtR. Multivariable Cox regression was used to estimate hazard ratios (HRs) and their 95% confidence intervals (CIs) for the incidence of CV outcomes. Model 1 was unadjusted. Model 2 was adjusted for age, sex, BMI, duration of diabetes, smoking status, alcohol consumption, and physical activity. Model 3 was further adjusted for SBP, HTN, dyslipidemia, LDL-C, FPG, and eGFR.
Subgroup stratified analyses were conducted to evaluate the interactions between the WHtR and the following variables: age, sex, BMI, smoking status, alcohol consumption, HTN, dyslipidemia, and CKD. Additionally, stratified analyses were conducted based on CO defined by WC (WC ≥90 cm for males and ≥85 cm for females) [32].
We used area under the receiver operating characteristic curve analysis to determine the optimal cutoff value and to assess the discriminative ability of the model in our study population.
All analyses were conducted using the R version 4.4.2 software (R Foundation for Statistical Computing, Vienna, Austria) and SAS Enterprise Guide 7.1 for Windows (SAS Institute, Cary, NC, USA). Statistical significance was set at a two-sided P value <0.05.
RESULTS
Baseline characteristics
In Table 1, the baseline characteristics of the study participants are presented according to WHtR categories using cutoffs of 0.5 and 0.6. The average age of the study population was 54.5 years, with a predominance of males (69.7%). BMI was 21.6±2.3, 25.3±2.5, and 30.0±3.6 kg/m² in the WHtR <0.5, 0.5 to <0.6, and ≥0.6 groups, respectively (P<0.001), and WC also increased with increasing WHtR (76.2±6.1, 88.2±5.8, and 100.2±7.5 cm; P<0.001). As WHtR increased, SBP, DBP, TC, TG, AST, and ALT all showed increasing trends, while HDL-C and eGFR decreased (all P<0.001).
Baseline characteristics according to WHtR categories using cutoffs of 0.5 and 0.6
Compared to those with WHtR <0.5 and 0.5 to <0.6 groups, individuals with WHtR ≥0.6 had lower rates of current smoking (15.2% vs. 38.1% and 28.1%) and heavy alcohol consumption (8.8% vs. 12.1% and 13.0%), but had a higher prevalence of physical inactivity (83.0% vs. 75.4% and 76.4%). Regarding comorbidities, the prevalence of HTN, dyslipidemia, and CKD was highest in the WHtR ≥0.6 group.
CV outcomes according to WHtR
During a median follow-up period of 6.7 years (interquartile range, 5.2 to 8.8), 4,293 primary composite CV events occurred. Fig. 2A reveals a stepwise increase in the risk of composite CV outcome across WHtR categories (<0.5, 0.5 to <0.6, and ≥0.6; P<0.001).
Cumulative incidence of the composite cardiovascular outcomes according to waist-to-height ratio (WHtR) categories. (A) T hree-group classification (<0.5, 0.5 to <0.6, and ≥0.6). (B) Two-group classification (<0.5 and ≥0.5). (C) Two-group classification (<0.6 and ≥0.6).
The incidence rate (per 1,000 person-years) of composite CV outcome was 29.85 in the WHtR <0.5 group, 42.41 in the 0.5 to <0.6 group and 55.80 in the ≥0.6 group (Table 2). After adjusting for age, sex, BMI, duration of diabetes, smoking status, alcohol consumption, physical activity, SBP, HTN, dyslipidemia, LDL-C, FPG, and eGFR, adjusted hazard ratio (aHR) for composite CV outcome consistently exhibited an increasing trend as WHtR rises. Compared to the WHtR <0.5 group, the aHR for composite CV outcome was 1.14 (95% CI, 1.05 to 1.24) in the WHtR 0.5 to <0.6 group and 1.62 (95% CI, 1.38 to 1.90) in the WHtR ≥0.6 group (P for trend <0.001).
Hazard ratios for the primary composite cardiovascular outcomes according to WHtR
The optimal cutoff for the composite CV outcome in our study population was 0.52 (Supplementary Table 1). When we classified WHtR into two groups using WHtR cutoff values of 0.5, individuals with WHtR ≥0.5 showed a greater risk of composite CV outcomes compared to those with WHtR <0.5 (aHR in model 3, 1.10 [95% CI, 1.02 to 1.20]) (Table 2, Fig. 2B). The association was stronger when a cutoff of 0.6 was applied: individuals with WHtR ≥0.6 had an aHR of 1.39 (95% CI, 1.23 to 1.57) compared to those with WHtR <0.6 (Table 2, Fig. 2C).
The detrimental effect of high WHtR was also evident across all components of the composite CV outcomes, including HF, MI, ischemic stroke, and CVD-related deaths (Table 3). A gradual increase in the risk of each CV outcome was noted with higher WHtR levels (0.5 to <0.6, and ≥0.6), using a WHtR <0.5 as the reference group. For HF, the aHRs were 1.07 (95% CI, 0.96 to 1.19) for the WHtR 0.5 to <0.6 group and 1.61 (95% CI, 1.32 to 1.96) for the WHtR ≥0.6 group (P for trend <0.001). For MI, the corresponding values were 1.25 (95% CI, 1.06 to 1.49) and 1.72 (95% CI, 1.24 to 2.37) (P for trend= 0.001), while for ischemic stroke, they were 1.15 (95% CI, 1.01 to 1.32) and 1.47 (95% CI, 1.14 to 1.90) (P for trend=0.005). In the case of CVD-related death, aHRs were 1.03 (95% CI, 0.79 to 1.34) for the WHtR 0.5 to <0.6 group and 2.16 (95% CI, 1.37 to 3.39) for the WHtR ≥0.6 group (P for trend=0.020).
Hazard ratios for the individual cardiovascular outcomes according to WHtR
Subgroup analysis
Subgroup analyses were conducted comparing WHtR ≥0.5 vs. <0.5 and WHtR ≥0.6 vs. <0.6 (Fig. 3). Regardless of whether the cutoff point was set at 0.5 or 0.6, higher WHtR was generally associated with a higher risk of the composite outcome across subgroups. In both analyses, a significant interaction was observed among the age subgroups, with a stronger association in individuals aged <65 years. In contrast, no significant interaction was found for sex. In the sex-stratified analyses (Supplementary Table 2), the risk of composite CV events increased with higher WHtR categories in both sexes. A statistically significant linear trend was observed across the three WHtR categories (<0.5, 0.5 to <0.6, ≥0.6), with P for trend <0.001 in males and 0.002 in females.
Subgroup analysis. Adjusted hazard ratios for the composite cardiovascular outcomes are presented, comparing the waist-to-height ratio (WHtR) ≥0.5 group to the WHtR <0.5 group, and WHtR ≥0.6 group to the WHtR <0.6 group. Adjusted for age, sex, body mass index (BMI), duration of diabetes, smoking status, alcohol consumption, physical activity, systolic blood pressure, hypertension (HTN), dyslipidemia, low-density lipoprotein cholesterol, fasting plasma glucose, and estimated glomerular filtration rate. HR, hazard ratio; CI, confidence interval; CKD, chronic kidney disease.
We also conducted stratified analyses according to BMI (<25 kg/m² vs. ≥25 kg/m²) (Supplementary Table 3) and WC (<90 cm vs. ≥90 cm for males; and <85 cm vs. ≥85 cm for females) (Supplementary Table 4). The number of individuals with WHtR ≥0.6 was small among those with BMI <25 kg/m² or those with WC <90 cm (males) or <85 cm (females), whereas the number of individuals with WHtR <0.5 was small among those with BMI ≥25 kg/m² or those with WC ≥90 cm (males) or ≥85 cm (females). Generally, higher WHtR was associated with an increased risk of composite CV outcomes. Among individuals with BMI <25 kg/m², compared to the WHtR <0.5 group, the aHRs for composite CV events were 1.25 (95% CI, 1.13 to 1.37) for the WHtR 0.5 to <0.6 group and 1.58 (95% CI, 1.06 to 2.36) for the WHtR ≥0.6 group (Supplementary Table 3). In those with BMI ≥25 kg/m², the corresponding values were 1.22 (95% CI, 0.94 to 1.58) and 1.40 (95% CI, 1.04 to 1.89), respectively. In WC-stratified analyses (Supplementary Table 4), WHtR ≥0.5 was associated with an increased risk of composite CV outcomes among individuals with WC <90 cm (males) or <85 cm (females). Among those with WC ≥90 cm (males) or ≥85 cm (females), WHtR ≥0.6 was significantly associated with increased risk compared to WHtR <0.6.
Two-time point analysis
The WHtR values measured at the two health checkups were categorized into CO+ (WHtR ≥0.6) or CO− (WHtR <0.6), with the results from the two checkups connected using an arrow (→). The characteristics of the four groups (CO− → CO−, CO− → CO+, CO+ → CO−, and CO+ → CO+) on the index date (the date of the latter health checkup) are presented in Supplementary Table 5. The (CO+ → CO+) group revealed the highest values for weight, BMI, and WC, followed by (CO+ → CO−), (CO− → CO+), and (CO− → CO−) groups. However, the (CO− → CO+) group had higher TG and lower HDL-C levels, higher AST and ALT levels, and a higher proportion of physical inactivity and CKD compared to the (CO+ → CO−) group.
To examine the association between changes in the WHtR and composite CV outcome, aHRs were estimated using the (CO− → CO−) group as the reference group (Supplementary Table 6). The (CO− → CO+) group and the (CO+ → CO+) group demonstrated an increased risk of composite CV outcome (aHR in model 3, 1.53 [95% CI, 1.10 to 2.13]) and 1.76 [95% CI, 1.34 to 2.31], respectively). Conversely, the (CO+ → CO−) group did not exhibit a statistically significant increase in risk (aHR, 1.22 [95% CI, 0.90 to 1.66]). When CO+ was defined as WHtR ≥0.5, none of the groups exhibited a significant increase in risk (Supplementary Table 7).
DISCUSSION
In this nationwide longitudinal study, a high WHtR was associated with an increased risk of CV outcomes, including HF, MI, ischemic stroke, and CVD-related death, among adults with T1DM. A clear trend of increasing CV risk with higher WHtR levels was noted when the WHtR was classified into three groups (<0.5, 0.5 to <0.6, and ≥0.6), and this pattern was consistent in both sexes.
T1DM is related to increased mortality compared with the general population, and CVD is the leading cause of these deaths [34,35]. As of 2021, an estimated 8.4 million people worldwide had T1DM, and this burden is expected to grow rapidly [36]. CO and MetS prevalence in T1DM is also rising, and this ‘double diabetes’ condition is known to exacerbate the risk of long-term complications of diabetes [1,10,37-40]. Despite the growing importance of weight management, attaining effective weight control in patients with T1DM remains more challenging than in the general population [41]. Thus, it is important to identify CO markers that can effectively predict the CV risk in patients with T1DM. The WHtR has been demonstrated to outperform other obesity measures, such as BMI, WC, and waist-to-hip ratio, in predicting CV risk in the general population and those with T2DM [15,42,43]. However, evidence linking the WHtR to CV risk in patients with T1DM remains limited.
Prior research has established an association between WHtR ≥0.5 and diabetic complications, including diabetic eye disease and HF in T1DM [18,44]. However, to the best of our knowledge, this is the first nationwide cohort study to explore the association between the WHtR and composite outcomes of comprehensive CV risk. Moreover, instead of the single threshold of 0.5, we employed two cutoff values of 0.5 and 0.6 and revealed a clear, sequential increase in the CV risk as the WHtR levels increased across the three groups.
Furthermore, our study has strengths in that we evaluated the impact of changes in the WHtR on the CV risk by conducting a two-point analysis. Previous studies have typically measured the WHtR only once at baseline to analyze its association with outcomes. Consequently, little is known regarding how changes in the WHtR over time impact the CV risk. In our study, compared to the (CO− → CO−) group, not only the (CO+ → CO+) but also the (CO− → CO+) group showed a significant increase in HR, indicating that maintaining a WHtR <0.6 is critical. On the other hand, the (CO+ → CO−) group did not exhibit an increase in HR, indirectly suggesting that reducing visceral fat is important for individuals with a WHtR ≥0.6 to prevent CVD in T1DM. However, owing to the limited sample size, this study was unable to compare the (CO+ → CO+) and (CO+ → CO−) groups and directly demonstrate the protective effects of WHtR reduction on CVD. When CO+ was defined as WHtR ≥0.5, none of the groups exhibited a significant increase in risk. It is possible that the presence or absence of a more definitive shift toward CO better distinguishes CV risk; however, further studies in larger populations are required to draw firm conclusions and overcome current limitations.
At the baseline visit, the groups with an elevated WHtR showed a higher prevalence of metabolic disorders, including HTN, dyslipidemia, and CKD, indicating that the WHtR is a possible marker for identifying metabolic comorbidities in T1DM. Additionally, these groups had lower smoking and alcohol consumption rates but higher physical inactivity. This is likely attributable to the larger proportion of women in the elevated WHtR group than in the lower WHtR group, as women in South Korea typically display such patterns [45]. Importantly, the association between the WHtR and CV outcomes was significant after adjusting for these confounding variables.
Furthermore, the association persisted after adjusting for BMI, emphasizing the independent role of WHtR in predicting CV risk. Interestingly, a significant association between the WHtR and composite CV outcomes was noted in individuals with obesity (BMI ≥25 kg/m2). These findings suggest that the WHtR can differentiate CV risk even in individuals with obesity. Additionally, WHtR demonstrated better discriminatory ability for composite CV outcomes than WC (P<0.001) (Supplementary Table 8). In stratified analyses based on CO defined by WC (WC ≥90 cm for males and ≥85 cm for females), a high WHtR (WHtR ≥0.6 in individuals with WC-defined CO and WHtR ≥0.5 in those without) was significantly associated with an increased risk of composite CV outcomes, suggesting that incorporating WHtR may improve the identification of high-risk individuals beyond WC alone.
In the subgroup analysis, the impact of a high WHtR on composite CV outcome was greater in individuals <65 years of age than in those aged ≥65 years. Similarly, in the general population, prior research found the association between WHtR and CVD risk to be stronger in individuals aged <70 years than those aged ≥70 years [46]. The attenuated effect of the WHtR in the older age group was likely due to the dominant influence of age-related factors on CVD risk, such as endothelial dysfunction and arterial stiffness [47]. Still, older adults aged ≥65 years with WHtR ≥0.6 also exhibited a significantly increased risk of composite CV outcomes.
This study has certain limitations. First, data on glycosylated hemoglobin levels, a key factor in diabetic complications, were not available in the dataset. We used FPG as a substitute for adjustment, which may not fully reflect the glycemic status. Second, our dataset lacked information on the age at diabetes onset for individuals with a diabetes duration of >3 years. Consequently, we could only categorize the diabetes duration as <3 years or longer. Third, detailed information on insulin treatment, such as total daily dosage or the type of insulin, was not available in our dataset. Fourth, we were unable to incorporate data from the Type 1 Diabetes Registration Project for the Reimbursement of Consumable Materials, as it is not included in the standard NHIS database. Instead, we defined T1DM using an internally validated algorithm based on diagnostic codes and insulin prescription patterns, which has demonstrated high accuracy, with a positive predictive value of 95% and a sensitivity of 90% [26]. Indeed, the number of participants with T1DM who received a health examination in our study is comparable to the estimated T1DM population adjusted for the general health examination rate [48-50]. Furthermore, the retrospective nature of this study has inherent limitations, including potential residual confounding and the inability to infer causality. Lastly, as this study was limited to a single ethnic group, further studies involving diverse populations are required.
In this large nationwide longitudinal cohort study, a high WHtR was associated with an increased risk of CV complications—including HF, MI, ischemic stroke, and CVD-related death—in adults with T1DM. These findings highlight that the WHtR is a practical marker for assessing the CV risk in patients with T1DM. Measuring WHtR in routine clinical practice may aid in the early detection of high-risk individuals. When encountering patients with an elevated WHtR, clinicians should assess for coexisting CVD and manage CV risk factors more thoroughly, including lifestyle interventions such as weight management, to prevent CVD.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2025.0179.
Optimal WHtR cutoff values and AUROC for composite cardiovascular outcome prediction
Hazard ratios for composite cardiovascular outcomes according to WHtR, stratified by sex
Hazard ratios for composite cardiovascular outcomes according to WHtR, stratified by body mass index
Hazard ratios for the composite cardiovascular outcomes according to WHtR, stratified by waist circumference
Baseline characteristics according to WHtR change patterns between two-time points
Hazard ratios for composite cardiovascular outcomes according to WHtR change patterns between the two-time points, using a WHtR cutoff value as 0.6
Hazard ratios for composite cardiovascular outcomes according to WHtR change patterns between the two-time points, using a WHtR cutoff value as 0.5
AUROC-based comparison of WHtR and WC for composite cardiovascular outcome prediction
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: K.J.K., S.K., J.Y.K., J.H.K.
Acquisition, analysis, or interpretation of data: K.J.K., S.K., J.Y.K., J.H.K.
Drafting the work or revising: all authors.
Final approval of the manuscript: all authors.
FUNDING
None
ACKNOWLEDGMENTS
Additional data are available through the approval and oversight of the Korean National Health Insurance Service (https://nhiss.nhis.or.kr).