Abstract
Background
Metabolic syndrome (MetS) is a cluster of risk factors that increases cardiometabolic disease and mortality. A continuous MetS Score has been developed to quantify MetS severity, but its association with all-cause mortality in older Chinese adults remains unclear.
Methods
We analyzed 2,443 participants from the Chinese Longitudinal Healthy Longevity Survey (CLHLS, 2008–2018 and 2011–2018). MetS Score was derived by confirmatory factor analysis using triglycerides, HDL-C, fasting glucose, systolic blood pressure, and BMI. Participants were categorized into quartiles (Q1-Q4). Cox proportional hazards models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause mortality. Subgroup analyses examined heterogeneity by age, sex, residence, and other covariates, and restricted cubic splines explored dose–response relationships.
Results
During 10,356 person-years of follow-up, 1,412 deaths occurred (136.3/1,000 person-years). Each 1-unit increase in MetS Score was associated with a 13.9% lower mortality risk (HR = 0.861, 95% CI: 0.787–0.942). The protective effect was concentrated in Q4 vs. Q1 (HR = 0.792, 95% CI: 0.669–0.938). Subgroup analyses showed stronger associations in those aged ≥80 years (HR = 0.785, 95% CI: 0.660–0.935), women (HR = 0.758, 95% CI: 0.617–0.931) and rural residents (HR = 0.817, 95% CI: 0.618–0.980). Restricted cubic splines confirmed a linear inverse association overall and among the ≥80 years group.
Conclusion
MetS Score, reflecting MetS severity, was inversely associated with all-cause mortality in older Chinese adults, particularly among those aged ≥80 years. These findings suggest that higher MetS Score may paradoxically confer survival benefits in the older adults, warranting further mechanistic studies.
Keywords: all-cause mortality, Chinese older population, CLHLS, metabolic syndrome, MetS Score
Introduction
Metabolic syndrome (MetS), defined as a cluster of cardiometabolic abnormalities including central obesity, insulin resistance, hypertension, and dyslipidemia, has become a major global public health concern (1, 2). Substantial evidence indicates that both MetS and its individual components independently increase the risk of cardiovascular diseases (CVDs) and related adverse outcomes (3–5). A recent study demonstrated that the global prevalence of MetS increased from 11.9% in 2000 to 28.4% in 2023, affecting approximately 1.54 billion adults worldwide (6). In China Mainland, the prevalence of MetS also has increased strikingly to 24.5% in two decades (7). This striking increasing trend is mainly attributed to population aging, urbanization, and lifestyle changes associated with rapid economic development (6, 7). These findings highlight the urgent need for specific assessment strategies for metabolic syndrome in aging populations.
Conventional definitions, such as the NCEP ATP III criteria (8), classify MetS when three or more abnormalities are present. While clinically straightforward, this approach cannot capture the severity of metabolic dysfunction. To address this limitation, researchers have proposed continuous severity scores (“MetS Score”) (9–11), derived through confirmatory factor analysis of triglycerides, HDL-C, fasting glucose, blood pressure, and adiposity indices. These scores provide a graded measure of risk and outperform categorical definitions in predicting diabetes, CVD events, and related biomarkers (12–14). Notably, a recent Chinese study developed an age-, sex-, and ethnicity-specific MetS Score, which showed strong associations with CVD risk factors and biomarkers (15).
Despite these advances, the relationship between MetS Score and all-cause mortality remains unclear, especially in older adults. Some studies report linear associations, whereas others suggest paradoxical or J-shaped patterns, possibly reflecting the “obesity paradox” and age-related changes in body composition (16, 17). These uncertainties highlight the importance of evaluating MetS Score in China’s rapidly aging population.
Using data from the Chinese Longitudinal Healthy Longevity Survey (CLHLS), we aimed to assess the association between MetS Score and all-cause mortality in older Chinese adults, with further evaluation of subgroup differences, particularly among those aged ≥80 years, and the potential linearity of this relationship.
Methods
Study population
The CLHLS is a nationally representative, prospective cohort of adults aged ≥65 years across 23 provinces of China (18). More detailed information of CLHLS study design can be found at the following website: https://www.icpsr.umich.edu/web/NACDA/studies/38899 (19).
All interviewees were informed and signed the consent form willingly before collecting blood samples and conducting physical examinations. Ethical approval was obtained from the Biomedical Ethics Committee of Peking University (IRB00001052–13074), and all participants provided written informed consent.
For the present analysis, consistent with previous studies using the CLHLS data, older adults were defined as those aged 65 years and above (18). we used three datasets: the CLHLS biomarker dataset, and two follow-up datasets (2008–2018 and 2011–2018). Participants were excluded if they were <65 years of age, lacked biomarker measurements, were duplicated across survey waves, had no survival outcome data, or were lost to follow-up before the subsequent wave. After exclusions, 2,443 participants were eligible and included in the analysis. A detailed flowchart of enrollment and exclusions is presented in Figure 1.
Figure 1.
The flowchart of participants enrollment from CLHLS.
Data collection
Biological specimens were collected according to standardized protocols, and blood samples were processed and analyzed in certified central laboratories using calibrated equipment (20). Assays included triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), fasting blood glucose (FBG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C).
Physical examinations were conducted by trained physicians from the local Centers for Disease Control (CDC) (19). Measurements comprised systolic blood pressure (SBP), diastolic blood pressure (DBP), height, and weight, from which body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared.
Information on sociodemographic characteristics and lifestyle factors was obtained through structured face-to-face interviews (19). Covariates included age (<80 vs. ≥80 years), sex, ethnicity (Han vs. minority), residence (urban vs. rural), education (illiterate, primary, middle school, high school or above), marital status (married/cohabiting vs. other), living arrangement (alone, institution, or with family), pension status (yes vs. no), smoking (never, former, current), alcohol consumption (never, former, current), and physical activity (never, former, current). The cognitive function of participants at baseline were measured using the Chinese version of the Mini-Mental State Examination (CMMSE) and were catergorized into 2 groups accordding to previous studies (21, 22). Estimated glomerular filtration rate (eGFR) was estimated according to the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and catergoried into low eGFR group (< 60 mL/min per 1.73 m2) and high eGFR group (≥ 60 mL/min per 1.73 m2) (23). The activities of daily living (ADL) disability score, constructed by six daily activities items, including feeding, bathing, dressing, toileting, indoor transferring and continence, were used to measure ADL disability refer to previously studies. The ADL disability score ranged from 0 to 6 points (21, 24). The numbers of chronic diseases of participants were obtained by asking whether they had been diagnosed with hypertension, diabetes mellitus, heart disease, stroke, bronchitis or asthma and cancer.
Calculation of the MetS Score
A continuous MetS Score was constructed using confirmatory factor analysis (CFA) of five continuous metabolic indicators: TG, HDL-C, FBG, SBP, and BMI (9, 15, 25). Specifically, the skewed indicators including TG, FBG, and BMI were natural log-transformed and HDL-C values was inverted so that higher values of all indicators consistently indicated greater metabolic risk. All five indicators were then standardized and used to estimate the factor loadings in a single-factor CFA model. Finally, following the method of Gurka et al. (25), the final individual MetS Scores equation were derived from the standardized CFA model by back-transforming the estimated coefficients.
Clinical outcomes
The primary outcome was all-cause mortality, determined from CLHLS follow-up surveys conducted between 2008–2018 and 2011–2018. Vital status and dates of death were obtained during household interviews with close family members or village doctors and, when available, verified against official registration records.
Statistical analysis
Participants were categorized into quartiles of the MetS Score (Q1-Q4). Baseline characteristics were summarized as mean ± standard deviation (SD) or median (interquartile range, IQR) for continuous variables and as proportions for categorical variables. Differences across quartiles were assessed using analysis of variance (ANOVA) or Kruskal-Wallis tests for continuous variables and χ2 tests for categorical variables.
The association between MetS Score and all-cause mortality was evaluated using Cox proportional hazards regression, with hazard ratios (HRs) and 95% confidence intervals (CIs) reported. The proportional hazards assumption was tested using Schoenfeld residuals, and no significant violations were observed. Three nested models were specified: Model 1, crude and unadjusted; Model 2, adjusted for age, sex, and ethnicity; and Model 3, further adjusted for residence place, coresidence type, education, marital status, pension status, smoking, drinking, exercising status, survey year, ADL disability score, cognitive function status, eGFR and the number of chronic diseases. Linear trends across quartiles were tested by modeling the median value of each quartile as a continuous variable, as described in previous studies (26).
Prespecified subgroup analyses were conducted by age (<80 vs. ≥80 years), sex, residence place, coresidence type, marital status, pension status, education status, cognitive function status and status of chronic diseases. Restricted cubic spline regression was applied using the MetS Score as a continuous variable, with knots placed at the 5th, 35th, 65th, and 95th percentiles (27). Departure from linearity was tested using likelihood ratio tests.
Missing covariates were addressed using multilevel multiple imputation with chained equations and random effects, accounting for the hierarchical structure of the CLHLS survey data (28). Twenty imputed datasets were created, and final estimates were pooled according to Rubin’s rules (29, 30). To address potential reverse causation, we conducted sensitivity analyses by excluding participants who died within the first year and within the first 2 years of follow-up. All analyses were performed using R software (version 4.4.3), which is open-source and freely available.1 A two-sided p < 0.05 was considered statistically significant.
Results
Development of the MetS Score
Using CFA, a continuous MetS Score was derived from five metabolic indicators (TG, HDL-C, FBG, SBP, and BMI). The fitted formula for participants aged ≥65 years was: MetS Score = −2.095 + 0.003 * SBP + 0.501 * ln(BMI) + 0.673 * ln(FBG) -0.664 * HDL-C + 0.793 * ln(TG).
Factor loadings were highest for TG (0.569), followed by HDL-C (0.394), FBG (0.363), BMI (0.179), and SBP (0.104) (Supplementary Table S1).
Baseline characteristics of participants
Baseline characteristics of participants across quartiles of MetS Score are summarized in Table 1. Among 2,443 participants, 58.7% were female, and 71.9% were aged ≥80 years. Most were Han (91.2%), rural residents (82.2%), lived with family or in institutions (80.5%), had no pension (92.1%), illiterate (67.50%), had normal cognitive function (CMMSE score ≥18, 76.0%), and had no chronic diseases (69.3%). Nearly 17.7% were current smokers, 17.3% current drinkers, and 16.7% engaged in regular exercise.
Table 1.
Baseline characteristics of participants from CLHLS according to quartiles of MetS Score.
| Characteristics | Total (n = 2,443) | Q1 (<−0.44) (n = 610) | Q2 (−0.44to −0.07) (n = 609) | Q3 (−0.06–0.35) (n = 612) | Q4 (≥0.36) (n = 612) | p-value |
|---|---|---|---|---|---|---|
| Age (years), median (IQR) | 88 (22) | 89 (21) | 91 (20) | 91 (20) | 83 (21) | <0.001 |
| Age (years) | <0.001 | |||||
| <80 | 686 (28.08) | 157 (25.74) | 142 (23.32) | 152 (24.84) | 235 (38.40) | |
| ≥80 | 1757 (71.92) | 453 (74.26) | 467 (76.68) | 460 (75.16) | 377 (61.60) | |
| Sex | <0.001 | |||||
| Male | 1,010 (41.34) | 290 (47.54) | 218 (35.80) | 232 (37.91) | 270 (44.12) | |
| Female | 1,433 (58.66) | 320 (52.46) | 391 (64.20) | 380 (62.09) | 342 (55.88) | |
| Ethnicity | 0.004 | |||||
| Han nationality | 2,227 (91.16) | 576 (94.43) | 558 (91.63) | 546 (89.22) | 547 (89.38) | |
| Minority nationality | 216 (8.84) | 34 (5.57) | 51 (8.37) | 66 (10.78) | 65 (10.62) | |
| Marital status | <0.001 | |||||
| Married/cohabiting | 807 (33.03) | 221 (36.23) | 170 (27.91) | 170 (27.78) | 246 (40.20) | |
| Divorced/never married/separated/widowed | 1,636 (66.97) | 389 (63.77) | 439 (72.09) | 442 (72.22) | 366 (59.80) | |
| Residence place | <0.001 | |||||
| Urban | 435 (17.81) | 93 (15.25) | 87 (14.29) | 87 (14.21) | 168 (27.45) | |
| Rural | 2008 (82.19) | 517 (84.75) | 522 (85.71) | 525 (85.78) | 444 (72.55) | |
| Coresidence type | 0.828 | |||||
| Alone | 477 (19.53) | 127 (20.82) | 116 (19.05) | 118 (19.28) | 116 (18.95) | |
| Not alone (lived with family members or in an institute) | 1966 (80.47) | 483 (79.18) | 493 (80.95) | 494 (80.72) | 496 (81.05) | |
| Education levels | 0.009 | |||||
| Illiterate | 1,649 (67.50) | 399 (65.41) | 439 (72.09) | 434 (70.92) | 377 (61.60) | |
| Primary school | 623 (25.50) | 168 (27.54) | 133 (21.84) | 142 (23.20) | 180 (29.41) | |
| Middle school | 112 (4.58) | 30 (4.92) | 22 (3.61) | 24 (3.92) | 36 (5.88) | |
| High school and above | 59 (2.42) | 13 (2.13) | 15 (2.46) | 12 (1.96) | 19 (3.10) | |
| Smoking status | 0.806 | |||||
| Never | 1800 (73.68) | 441 (72.30) | 464 (76.19) | 445 (72.71) | 450 (73.53) | |
| Former | 211 (8.64) | 57 (9.34) | 46 (7.55) | 55 (8.99) | 53 (8.66) | |
| Current | 432 (17.68) | 112 (18.36) | 99 (16.26) | 112 (18.30) | 109 (17.81) | |
| Drinking status | 0.158 | |||||
| Never | 1872 (76.63) | 453 (74.26) | 464 (76.19) | 480 (78.43) | 475 (77.61) | |
| Former | 149 (6.10) | 35 (5.74) | 32 (5.25) | 37 (6.05) | 45 (7.35) | |
| Current | 422 (17.27) | 122 (20.00) | 113 (18.56) | 95 (15.52) | 92 (15.03) | |
| Exercise | <0.001 | |||||
| Never | 1919 (78.55) | 509 (83.44) | 492 (80.79) | 478 (78.10) | 440 (71.90) | |
| Former | 115 (4.71) | 18 (2.95) | 30 (4.93) | 36 (5.88) | 31 (5.07) | |
| Current | 409 (16.74) | 83 (13.61) | 87 (14.29) | 98 (16.01) | 141 (23.04) | |
| BMI | <0.001 | |||||
| Underweight | 742 (30.37) | 245 (40.16) | 197 (32.35) | 180 (29.41) | 120 (19.61) | |
| Normal | 1,313 (53.75) | 327 (53.61) | 334 (54.84) | 329 (53.76) | 323 (52.78) | |
| Overweight and obesity | 388 (15.88) | 38 (6.23) | 78 (12.81) | 103 (16.83) | 169 (27.61) | |
| Pension subsidies | 0.001 | |||||
| No | 2,251 (92.14) | 559 (91.64) | 572 (93.92) | 577 (94.28) | 543 (88.73) | |
| Yes | 192 (7.86) | 51 (8.36) | 37 (6.08) | 35 (5.72) | 69 (11.27) | |
| Survey year | ||||||
| 2008 | 1,123 (46.0) | 86 (14.1) | 281 (46.1) | 367 (60.0) | 389 (63.6) | <0.001 |
| 2012 | 1,320 (54.0) | 524 (85.9) | 328 (53.9) | 245 (40.0) | 223 (36.4) | |
| CMMSE score | ||||||
| <18 | 587 (24.0) | 130 (21.3) | 182 (29.9) | 167 (27.3) | 108 (17.6) | <0.001 |
| ≥ 18 | 1856 (76.0) | 480 (78.7) | 427 (70.1) | 445 (72.7) | 504 (82.4) | |
| eGFR (ml/min per 1.73 m2) | ||||||
| Low group (<60) | 757 (31.0) | 144 (23.6) | 177 (29.1) | 213 (34.8) | 223 (36.4) | <0.001 |
| High group (≥ 60) | 1,686 (69.0) | 466 (76.4) | 432 (70.9) | 399 (65.2) | 389 (63.6) | |
| Numbers of the self-reported chronic diseases | ||||||
| 0 | 1,693 (69.3) | 436 (71.5) | 422 (69.3) | 424 (69.3) | 411 (67.2) | 0.18 |
| 1 ~ 2 | 724 (29.6) | 169 (27.7) | 181 (29.7) | 185 (30.2) | 189 (30.9) | |
| ≥ 3 | 26 (1.1) | 5 (0.8) | 6 (1.0) | 3 (0.5) | 12 (2.0) | |
| The ADL disability score | 0.59 (1.44) | 0.64 (1.47) | 0.65 (1.50) | 0.61 (1.42) | 0.45 (1.36) | 0.062 |
MetS, metabolic syndrome; Q1–Q4, quartile 1-quartile 4; IQR, interquartile range; BMI, body mass index. ADL, activities of daily living; CMMSE, the Chinese version of the Mini Mental State Examination; eGFR, Estimated glomerular filtration rate.
Quartile 1: MetS<−0.44, Quartile 2: −0.44 < MetS<−0.07, Quartile 3: −0.06 < MetS<0.35, Quartile 4: MetS≥0.36.
Median (IQR) MetS Scores were −0.70 (0.37) for Q1, −0.24 (0.18) for Q2, 0.14 (0.20) for Q3, and 0.75 (0.63) for Q4. Participants in Q4 were younger, more likely to be female, urban residents, married or cohabiting, overweight/obese, and more likely to receive pensions, engage in physical activity, with lower eGFR and normal cognitive function (CMMSE score ≥18).
Association between the MetS Score and all-cause mortality
Kaplan–Meier survival curves illustrating mortality across quartiles of MetS Score are shown in Figure 2. During 10,356.6 person-years of follow-up (median 3.8 years), 1,412 deaths occurred, corresponding to a crude mortality rate of 136.3 per 1,000 person-years. Mortality rates by quartile were 145.2, 148.8, 153.8, and 103.1 per 1,000 person-years for Q1-Q4, respectively. Kaplan–Meier curves demonstrated significant differences across quartiles (P for-log-rank = 0.001).
Figure 2.
The Kaplan–Meier survival analysis curve for all-cause mortality according to quartiles of MetS Score.
Table 2 presents the results of Cox proportional hazards regression models examining the association between MetS Score and all-cause mortality. Each one-unit increase in MetS Score was associated with a 19.6% reduction in mortality in the crude model (HR = 0.804, 95% CI: 0.743–0.869, p < 0.001). After adjustment for age, sex, and ethnicity, the association remained significant though attenuated (HR = 0.864, 95% CI: 0.796–0.938, p < 0.001). In the fully adjusted model, the protective association persisted (HR = 0.861, 95% CI: 0.787–0.942, p = 0.001).
Table 2.
The hazard ratios for the association between MetS Score and all-cause mortality (N = 2,443).
| Models | HR (95%CI) | p-value | HR (95%CI) | P for trend | |||
|---|---|---|---|---|---|---|---|
| As continuous variable | Q1 | Q2 | Q3 | Q4 | |||
| Model 1 | 0.804 (0.743–0.869) | <0.001 | Ref | 1.023 (0.883–1.186) | 1.056 (0.913–1.222) | 0.707 (0.605–0.828) | <0.001 |
| Model 2 | 0.864 (0.796–0.938) | <0.001 | Ref | 0.968 (0.835–1.123) | 1.026 (0.886–1.187) | 0.804 (0.687–0.942) | 0.012 |
| Model 3 | 0.861 (0.787–0.942) | 0.001 | Ref | 0.939 (0.804–1.096) | 0.979 (0.835–1.149) | 0.792 (0.669–0.938) | 0.009 |
MetS, metabolic syndrome; HR, hazard ratios; 95% CI, 95% confidence interval; ref, reference.
Model 1: crude model, no co-variables was adjusted; Model 2: adjusted for age, sex and ethnicity; Model 3: further adjusted for residence area, coresidence type, educational levels, current marital status, pension subsidies, smoking status, drinking status, exercising status, survey year, ADL disability score, cognitive function, eGFR and the numbers of chronic diseases.
When analyzed by quartiles, the highest quartile (Q4) demonstrated a consistent protective effect compared with Q1. The hazard ratios for Q4 were 0.707 (95% CI: 0.605–0.828, p < 0.001) in Model 1, 0.804 (95% CI: 0.687–0.942, p = 0.012) in Model 2, 0.792 (95% CI: 0.669–0.938, p = 0.009) in Model 3 (Table 2). Further sensitivity analyses in 2219 older adults demonstrated consistent results, confirming the stability of the observed associations (Supplementary Table S2). Kaplan–Meier survival curves (Figure 2) further illustrated the lower cumulative mortality risk in Q4 compared with the other quartiles. Restricted cubic spline analysis (Figure 3) confirmed a linear inverse dose–response association between MetS Score and mortality (P-overall = 0.001, P-nonlinear = 0.054).
Figure 3.
Restricted cubic spline model of the association between MetS Score and all-cause mortality in overall population.
Subgroup analysis
Tables 3, 4 summarize the subgroup analyses of the association between MetS Score and all-cause mortality. As a continuous variable, the MetS Score was significantly protective in several subgroups: those aged ≥80 years (HR = 0.847, 95% CI: 0.769–0.933, p < 0.001), women (HR = 0.839, 95% CI: 0.745–0.944, p = 0.003), unmarried individuals (divorced/never married/separated/widowed) (HR = 0.835, 95% CI: 0.753–0.927, p < 0.001), rural residents (HR = 0.870, 95% CI: 0.784–0.966, p = 0.009), those lived with family members or in institutions (HR = 0.847, 95% CI: 0.767–0.935, p < 0.001), those without pension subsidies(HR = 0.848, 95% CI: 0.772–0.931, p < 0.001), those with CMMSE score <18 (HR = 0.709, 95% CI: 0.592–0.848, p < 0.001), those with eGFR ≥60 mL/min per 1.73 m2 (HR = 0.848, 95% CI: 0.747–0.962, p = 0.010) and those without self-reported chronic diseases (HR = 0.784, 95% CI: 0.697–0.881, p < 0.001). Protective effects were evident in both illiterate participants (HR = 0.879, 95% CI: 0.791–0.978, p = 0.018) and those who received formal education (HR = 0.813, 95% CI: 0.675–0.978, p = 0.028).
Table 3.
Subgroup analysis of the association between MetS Score (as continuous variable) and the risk of all-cause mortality.
| Characteristics | Event/no. | As continuous variable | p-value |
|---|---|---|---|
| HR (95%CI) | |||
| Age (years) | |||
| <80 | 131/686 | 0.836 (0.638–1.096) | 0.195 |
| ≥80 | 1281/1757 | 0.847 (0.769–0.933) | <0.001 |
| Sex | |||
| Male | 499/1010 | 0.907 (0.783–1.051) | 0.194 |
| Female | 913/1433 | 0.839 (0.745–0.944) | 0.003 |
| Marital status | |||
| Married/cohabiting | 259/807 | 0.958 (0.789–1.161) | 0.659 |
| Divorced/never married/separated/widowed | 1153/1636 | 0.835 (0.753–0.927) | <0.001 |
| Residence place | |||
| Rural | 1172/2008 | 0.870 (0.784–0.966) | 0.009 |
| Urban | 240/435 | 0.820 (0.672–1.001) | 0.052 |
| Coresidence type | |||
| Alone | 260/477 | 0.892 (0.708–1.122) | 0.328 |
| Not alone | 1152/1966 | 0.847 (0.767–0.935) | <0.001 |
| Pension subsidies | |||
| No | 1341/2251 | 0.848 (0.772–0.931) | <0.001 |
| Yes | 71/192 | 1.175 (0.831–1.663) | 0.362 |
| Education levels | |||
| Illiterate | 1105/1649 | 0.879 (0.791–0.978) | 0.018 |
| Formal education | 307/794 | 0.813 (0.675–0.978) | 0.028 |
| CMMSE score | |||
| <18 | 516/587 | 0.709 (0.592–0.848) | <0.001 |
| ≥ 18 | 896/1856 | 0.927 (0.832–1.032) | 0.166 |
| eGFR (mL/min per 1.73 m2) | |||
| low group (<60) | 544/757 | 0.880 (0.770–1.005) | 0.060 |
| high group (≥ 60) | 868/1686 | 0.848 (0.747–0.962) | 0.010 |
| The self-reported chronic diseases | |||
| No | 985/1693 | 0.784 (0.697–0.881) | <0.001 |
| Yes | 427/750 | 0.984 (0.850–1.139) | 0.830 |
MetS, metabolic syndrome; HR, hazard ratios; 95% CI, 95% confidence interval; BMI, body mass index.
Model was adjusted for age, sex, ethnicity, residence area, coresidence type, educational levels, current marital status, pension subsidies, smoking status, drinking status, exercising status, survey year, ADL disability score, cognitive function, eGFR and the numbers of chronic diseases.
Table 4.
Subgroup analysis of the association between quartiles of MetS Score and the risk of all-cause mortality.
| Characteristics | Event/No. | Q1 | HR (95%CI) | P for trend | ||
|---|---|---|---|---|---|---|
| Q2 | Q3 | Q4 | ||||
| Age (years) | ||||||
| <80 | 131/686 | Ref | 1.348 (0.761–2.388) | 1.051 (0.578–1.913) | 1.033 (0.551–1.934) | 0.703 |
| ≥80 | 1281/1757 | Ref | 0.943 (0.801–1.111) | 0.919 (0.774–1.089) | 0.785 (0.660–0.935) | 0.005 |
| Sex | ||||||
| Male | 499/1010 | Ref | 1.080 (0.834–1.398) | 1.114 (0.848–1.462) | 0.939 (0.702–1.257) | 0.605 |
| Female | 913/1433 | Ref | 0.898 (0.742–1.087) | 0.931 (0.765–1.132) | 0.758 (0.617–0.931) | 0.012 |
| Marital status | ||||||
| Married/cohabiting | 259/807 | Ref | 1.064 (0.737–1.537) | 1.168 (0.796–1.713) | 1.046 (0.688–1.591) | 0.813 |
| Divorced/never married/separated/widowed | 1153/1636 | Ref | 0.954 (0.804–1.133) | 0.914 (0.765–1.092) | 0.798 (0.665–0.958) | 0.011 |
| Residence place | ||||||
| Rural | 1172/2008 | Ref | 0.922 (0.776–1.096) | 0.837 (0.700–1.000) | 0.817 (0.681–0.980) | 0.020 |
| Urban | 240/435 | Ref | 1.096 (0.737–1.630) | 1.436 (0.953–2.162) | 0.559 (0.336–0.930) | 0.055 |
| Coresidence type | ||||||
| Alone | 260/477 | 0.732 (0.494–1.084) | 0.828 (0.557–1.23) | 0.823 (0.544–1.245) | 0.627 | |
| Not alone (lived with family members or in an institute) | 1152/1966 | 1.030 (0.868–1.222) | 1.029 (0.862–1.228) | 0.791 (0.655–0.956) | 0.008 | |
| Pension subsidies | ||||||
| No | 1341/2251 | Ref | 0.978 (0.833–1.148) | 0.983 (0.833–1.160) | 0.793 (0.667–0.944) | 0.007 |
| Yes | 71/192 | Ref | 1.866 (0.883–3.941) | 2.905 (1.292–6.534) | 1.710 (0.714–4.096) | 0.178 |
| Education levels | ||||||
| Illiterate | 1105/1649 | Ref | 1.002 (0.839–1.195) | 0.993 (0.825–1.195) | 0.883 (0.731–1.066) | 0.157 |
| Formal education | 307/794 | Ref | 1.027 (0.736–1.432) | 1.005 (0.715–1.412) | 0.749 (0.509–1.103) | 0.130 |
| CMMSE score | ||||||
| <18 | 516/587 | Ref | 0.740 (0.573–0.955) | 0.731 (0.561–0.952) | 0.633 (0.483–0.829) | 0.002 |
| ≥ 18 | 896/1856 | Ref | 0.997 (0.82–1.212) | 1.110 (0.906–1.359) | 0.920 (0.741–1.144) | 0.518 |
| eGFR (mL/min per 1.73 m2) | ||||||
| Low group (<60) | 544/757 | Ref | 0.832 (0.647–1.068) | 1.101 (0.858–1.412) | 0.700 (0.536–0.915) | 0.032 |
| High group (≥ 60) | 868/1686 | Ref | 0.957 (0.787–1.164) | 0.969 (0.787–1.194) | 0.873 (0.701–1.088) | 0.245 |
| The self-reported chronic diseases | ||||||
| No | 985/1693 | Ref | 0.889 (0.738–1.072) | 0.911 (0.751–1.105) | 0.699 (0.569–0.859) | <0.001 |
| Yes | 427/750 | Ref | 1.118 (0.84–1.488) | 1.137 (0.846–1.529) | 1.029 (0.756–1.400) | 0.947 |
MetS, metabolic syndrome; HR, hazard ratios; 95% CI, 95% confidence interval; BMI, body mass index, ref, reference.
Model was adjusted for age, sex, ethnicity, residence area, coresidence type, educational levels, current marital status, pension subsidies, smoking status, drinking status, exercising status survey year, ADL disability score, cognitive function, eGFR and the numbers of chronic diseases.
When analyzed categorically, the protective effect remained consistent across most subgroups (Table 4). Compared with Q1, Q4 was significantly associated with lower mortality risk in participants aged ≥80 years (HR = 0.785, 95% CI: 0.660–0.935, P-for-trend = 0.005), women (HR = 0.758, 95% CI: 0.617–0.931, P-for-trend = 0.012), unmarried individuals(HR = 0.798, 95% CI: 0.665–0.958, P-for-trend = 0.011), rural residents (HR = 0.817, 95% CI: 0.618–0.980, P-for-trend = 0.020), those living with family members or in institutions(HR = 0.791, 95% CI:0.655–0.956, P-for-trend = 0.008), those without pension subsidies(HR = 0.793, 95% CI: 0.667–0.944, P-for-trend = 0.007) and those without chronic diseases (HR = 0.699, 95% CI: 0.569–0.859, P-for-trend <0.001).
In addition, restricted cubic spline analysis (Supplementary Figure S1) also confirmed the association between MetS Score and mortality among different subgroups.
Sensitivity and validation analyses
In sensitivity analyses, the inverse association between MetS Score and all-cause mortality remained consistent after excluding deaths within the first year and within the first 2 years of follow-up (Supplementary Tables S2, S3). Bootstrap internal validation showed stable CFA factor loadings and no substantial overfitting, with an optimism-corrected Harrell’s C-index similar to the apparent C-index (Supplementary Table S4). Incremental discrimination analyses showed that adding the MetS Score to the fully adjusted model produced a modest increase in Harrell’s C-index from 0.762 to 0.766 (Supplementary Table S5).
The associations between MetS Score and all-cause mortality in older population aged ≥80 years
Table 5 presents the results of Cox models restricted to the 1,757 participants aged ≥80 years. In this subgroup, each one-unit increase in MetS Score was associated with a 15.3% lower mortality risk in the fully adjusted model (HR = 0.847, 95% CI: 0.769–0.933, p < 0.001). Compared with Q1, those in Q4 consistently had lower mortality risk across all models: HR = 0.794 (95% CI: 0.677–0.932, P-for-trend = 0.003) in Model 1, HR = 0.794 (95% CI: 0.676–0.932, P-for-trend = 0.003) in Model 2, and HR = 0.785 (95% CI: 0.660–0.935, P-for-trend = 0.005) in Model 3.
Table 5.
Associations between MetS Score and the risk of all-cause mortality in adults aged 80 and above (N = 1757).
| Models | As continuous variable | p-value | Q1 | Q2 | Q3 | Q4 | P for trend |
|---|---|---|---|---|---|---|---|
| HR (95%CI) | HR (95%CI) | ||||||
| Model 1 | 0.841 (0.770–0.918) | <0.001 | Ref | 0.962 (0.823–1.124) | 0.925 (0.792–1.081) | 0.794 (0.677–0.932) | 0.003 |
| Model 2 | 0.840 (0.769–0.917) | <0.001 | Ref | 0.956 (0.817–1.118) | 0.920 (0.787–1.075) | 0.794 (0.676–0.932) | 0.003 |
| Model 3 | 0.847 (0.769–0.933) | <0.001 | Ref | 0.943 (0.801–1.111) | 0.919 (0.774–1.089) | 0.785 (0.660–0.935) | 0.005 |
MetS, metabolic syndrome; HR, hazard ratios; 95% CI, 95% confidence interval; ref, reference.
Model 1: crude model; Model 2:adjusted for sex and ethnicity; Model 3:further adjusted for residence area, coresidence type, educational levels, current marital status, pension subsidies, smoking status, drinking status, exercising status, survey year, ADL disability score, cognitive function, eGFR and the numbers of chronic diseases.
In addition, restricted cubic spline analysis (Supplementary Figure S1B) confirmed a linear inverse association between MetS Score and mortality among the oldest-old (P-overall = 0.002; P-nonlinear = 0.058).
Discussion
In this large, nationally representative cohort of older Chinese adults, we found that a higher MetS Score was inversely associated with all-cause mortality. This protective association was particularly evident among the oldest-old (≥80 years), women, unmarried participants and rural residents, and persisted after multivariable adjustment. When analyzed by quartiles, only participants in the highest quartile (Q4) consistently demonstrated lower mortality compared with Q1. Restricted cubic spline analyses confirmed a linear inverse relationship in the overall population and in the oldest-old subgroup.
Our findings provide new insights into the relationship between MetS severity and mortality in older adults. First, we observed that higher continuous MetS Scores were associated with lower all-cause mortality. This is in contrast to many studies conducted in middle-aged populations, where both traditional MetS definitions and continuous severity scores have been positively associated with incident CVD and mortality (12, 31). For example, analyses from the U.S. NHANES cohort and the Iranian Lipid and Glucose Study demonstrated that each unit increase in MetS severity score predicted higher risk of death and cardiovascular events, often with linear or J-shaped patterns (12, 32). Only participants in the highest quartile (Q4) showed a consistent survival benefit. In younger populations, mortality generally rises across the full gradient of MetS severity (33), whereas in our older sample the protective effect was limited to those with the greatest metabolic load. This pattern may reflect the “obesity paradox,” where greater adiposity and metabolic reserves become advantageous in late life (34). This may be because elevated TG and BMI reflect better nutritional status and energy reserves, protecting against malnutrition, sarcopenia, and frailty—key non-cardiovascular diseases mortality drivers in older adults (34–36).
Subgroup analyses showed stronger inverse associations among the oldest-old (≥80 years), women, and rural residents. Prior studies suggest that survival benefits of higher BMI and certain metabolic traits are more evident in women and the very old (16, 37), while urban–rural differences may reflect disparities in healthcare access and socioeconomic conditions. Our results extend recent work in China that developed an age-, sex-, and ethnicity-specific MetS Score and showed strong associations with CVD-related biomarkers (15). By linking this score to mortality outcomes, we provide novel evidence that the prognostic meaning of MetS severity differs by age, with paradoxical survival benefits emerging in the oldest-old.
Several mechanisms may underlie the paradoxical finding that higher MetS Scores were associated with lower mortality among older adults. One explanation is the well-described “obesity paradox,” whereby overweight or mildly obese individuals, especially in late life, demonstrate better survival than those with normal weight. Meta-analyses of patients with heart failure and other chronic conditions consistently show that higher BMI confers a survival advantage in older populations (38). A second explanation relates to nutritional and body-composition reserves: higher MetS Scores may partly capture greater fat and lean mass, which can provide essential energy and immunometabolic reserves during illness or stress. Epidemiologic studies indicate that sarcopenia is a strong predictor of mortality in older adults (39) and that inflammation-related muscle loss contributes to frailty and adverse outcomes (40). In this context, individuals with higher metabolic load but preserved muscle may have enhanced resilience compared with leaner counterparts. A third consideration is survivorship bias: those who reach advanced ages despite long-standing metabolic abnormalities may constitute a biologically robust subgroup with protective genetic, immunologic, or behavioral factors. Such selection effects have been widely discussed in epidemiologic methodology (41). Collectively, these mechanisms suggest that while high MetS severity increases cardiometabolic risk in younger populations, its prognostic meaning may shift in late life, where greater metabolic and nutritional reserves, combined with selective survival, could paradoxically indicate improved longevity.
Our findings have direct implications for geriatric risk assessment and management. In younger and middle-aged populations, continuous MetS Scores are valuable tools for identifying individuals at elevated cardiometabolic risk. However, in the oldest-old, their interpretation requires caution, as a higher score may not uniformly indicate vulnerability and may in some contexts reflect protective metabolic or nutritional reserves. These results underscore the importance of moving away from a “one-size-fits-all” threshold and adopting age-specific approaches to metabolic risk stratification. Clinicians should consider the broader physiological and social context of older adults when applying MetS-based tools, and policymakers should incorporate these nuances into guidelines for chronic disease prevention and healthy aging.
This study has several notable strengths, including the use of a large, nationally representative cohort of older Chinese adults, long-term follow-up with validated mortality outcomes, and application of a CFA-derived MetS Score that reflects metabolic severity on a continuous scale. Nonetheless, limitations should be acknowledged. First, residual confounding cannot be fully excluded despite extensive adjustment for demographic, socioeconomic, and lifestyle factors. Second, BMI was used as the adiposity measure; although practical, it does not distinguish between fat and lean mass, and future studies should incorporate waist circumference or imaging-based assessments of visceral fat. Third, the MetS Score was originally designed for cardiovascular risk prediction, and its applicability to all-cause mortality in adults aged ≥80 years may be limited, given that mortality in this population is driven by non-cardiovascular factors (e.g., infections, frailty, sarcopenia). Future studies are needed to validate the incremental predictive value of the MetS Score for all-cause mortality in the oldest-old. Fourth, the MetS Score was calculated using within-sample standardization of the five metabolic indicators rather than an external reference population, which may limit comparability with other studies using different reference standards or scoring methods. Fifth, selective survival bias and limited sample representativeness (2,443 participants, <25% of the original cohort) may restrict generalizability. Our findings are more likely to apply to relatively healthy, urban-dwelling older adults with higher socioeconomic status, rather than the general older population. Finally, cause-specific mortality could not be assessed, limiting our ability to determine whether the protective association was driven primarily by cardiovascular or non-cardiovascular deaths.
Conclusion
In conclusion, higher MetS Scores were paradoxically associated with lower all-cause mortality among older Chinese adults, particularly in the oldest-old subgroup. These findings highlight that the prognostic meaning of MetS severity differs across the life course and emphasize the need for age-tailored strategies in metabolic risk assessment and prevention.
Acknowledgments
The data and samples utilized in this research were obtained from the CLHLS. We extend our gratitude to the Peking University for their support.
Funding Statement
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by grants from the National Natural Science Foundation of China (Nos. 82574160, 82301768, and 82200312) and the startup fund for Principal Investigators from the School of Medicine, Shanghai Jiao Tong University (KJ2-0112-23-0002).
Edited by: Peter Piko, University of Debrecen, Hungary
Reviewed by: Ruixue Song, Shandong University, China
JiaHao Shi, Chulalongkorn University, Thailand
Abbreviations: MetS, Metabolic syndrome; IQR, Interquartile range; BMI, Body mass index; HR, Hazard ratio; ref, Reference.
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found at: https://www.icpsr.umich.edu/web/NACDA/studies/38899.
Ethics statement
Ethical approval was not required for the study involving humans in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was not required from the participants or the participants' legal guardians/next of kin in accordance with the national legislation and the institutional requirements.
Author contributions
CD: Conceptualization, Data curation, Methodology, Writing – original draft, Writing – review & editing. DH: Formal analysis, Writing – review & editing. JH: Conceptualization, Data curation, Methodology, Writing – original draft, Writing – review & editing. YX: Data curation, Writing – review & editing. HZ: Data curation, Writing – review & editing. YG: Formal analysis, Writing – review & editing. XZ: Formal analysis, Writing – review & editing. ZW: Writing – review & editing. XL: Conceptualization, Data curation, Formal analysis, Methodology, Supervision, Writing – original draft, Writing – review & editing.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpubh.2026.1832339/full#supplementary-material
(A–T) Restricted cubic spline models of the association between MetS Score and all-cause mortality in different predefined subgroups.
References
- 1.Saklayen MG. The global epidemic of the metabolic syndrome. Curr Hypertens Rep. (2018) 20:12. doi: 10.1007/s11906-018-0812-z, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Neeland IJ, Lim S, Tchernof A, Gastaldelli A, Rangaswami J, Ndumele CE, et al. Metabolic syndrome. Nat Rev Dis Primers. (2024) 10:77. doi: 10.1038/s41572-024-00563-5, [DOI] [PubMed] [Google Scholar]
- 3.Grundy SM. Metabolic syndrome update. Trends Cardiovasc Med. (2016) 26:364–73. doi: 10.1016/j.tcm.2015.10.004, [DOI] [PubMed] [Google Scholar]
- 4.Moghadam-Ahmadi A, Soltani N, Ayoobi F, Jamali Z, Sadeghi T, Jalali N, et al. Association between metabolic syndrome and stroke: a population based cohort study. BMC Endocr Disord. (2023) 23:131. doi: 10.1186/s12902-023-01383-6, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Haffner SM. Relationship of metabolic risk factors and development of cardiovascular disease and diabetes. Obesity (Silver Spring). (2006) 14:121S–7S. doi: 10.1038/oby.2006.291 [DOI] [PubMed] [Google Scholar]
- 6.Noubiap JJ, Nansseu JR, Nyaga UF, Ndoadoumgue AL, Ngouo AT, Tounouga DN, et al. Worldwide trends in metabolic syndrome from 2000 to 2023: a systematic review and modelling analysis. Nat Commun. (2025) 17:573. doi: 10.1038/s41467-025-67268-5, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Li R, Li W, Lun Z, Zhang H, Sun Z, Kanu JS, et al. Prevalence of metabolic syndrome in mainland China: a meta-analysis of published studies. BMC Public Health. (2016) 16:296. doi: 10.1186/s12889-016-2870-y, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute scientific statement. Circulation. (2005) 112:2735–52. doi: 10.1161/CIRCULATIONAHA.105.169404 [DOI] [PubMed] [Google Scholar]
- 9.Gurka MJ, Lilly CL, Oliver MN, DeBoer MD. An examination of sex and racial/ethnic differences in the metabolic syndrome among adults: a confirmatory factor analysis and a resulting continuous severity score. Metabolism. (2014) 63:218–25. doi: 10.1016/j.metabol.2013.10.006, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wijndaele K, Beunen G, Duvigneaud N, Matton L, Duquet W, Thomis M, et al. A continuous metabolic syndrome risk score: utility for epidemiological analyses. Diabetes Care. (2006) 29:2329. doi: 10.2337/dc06-1341, [DOI] [PubMed] [Google Scholar]
- 11.Honarvar M, Masoumi S, Mehran L, Khalili D, Amouzegar A, Azizi F. Development and validation of a continuous metabolic syndrome severity score in the Tehran lipid and glucose study. Sci Rep. (2023) 13:7529. doi: 10.1038/s41598-023-33294-w, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Honarvar M, Mehran L, Masoumi S, Agahi S, Khalili S, Azizi F, et al. Independent association between age- and sex-specific metabolic syndrome severity score and cardiovascular disease and mortality. Sci Rep. (2023) 13:14621. doi: 10.1038/s41598-023-41546-y, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schlesinger N, Elsaid MI, Rustgi VK. The relationship between metabolic syndrome severity and the risk of mortality in gout patients: a population-based study. Clin Exp Rheumatol. (2022) 40:631–3. doi: 10.55563/clinexprheumatol/2rn9fv, [DOI] [PubMed] [Google Scholar]
- 14.Elsaid MI, Bridges JFP, Li N, Rustgi VK. Metabolic syndrome severity predicts mortality in nonalcoholic fatty liver disease. Gastro Hep Adv. (2022) 1:445–56. doi: 10.1016/j.gastha.2022.02.002, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Yang S, Yu B, Yu W, Dai S, Feng C, Shao Y, et al. Development and validation of an age-sex-ethnicity-specific metabolic syndrome score in the Chinese adults. Nat Commun. (2023) 14:6988. doi: 10.1038/s41467-023-42423-y, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Flegal KM, Kit BK, Orpana H, Graubard BI. Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. (2013) 309:71–82. doi: 10.1001/jama.2012.113905, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Winter JE, MacInnis RJ, Wattanapenpaiboon N, Nowson CA. BMI and all-cause mortality in older adults: a meta-analysis. Am J Clin Nutr. (2014) 99:875–90. doi: 10.3945/ajcn.113.068122, [DOI] [PubMed] [Google Scholar]
- 18.Qi R, Yang Y, Sheng B, Li H, Zhang X. Plant-based diet indices and their association with frailty in older adults: a CLHLS-based cohort study. Nutrients. (2023) 15:5120. doi: 10.3390/nu15245120, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zeng Y, Shi X, Zheng Z, Lei X. Chinese Longitudinal Healthy Longevity Survey (CLHLS), [China], 1998–2018. Beijing: Inter-University Consortium for Political and Social Research; [Distributor] (2024). [Google Scholar]
- 20.Zeng Y, Vaupel JW. Chinese Longitudinal Healthy Longevity Survey (CLHLS), Biomarkers Datasets, 2009, 2012, 2014. Beijing: Inter-University Consortium for Political and Social Research, [Distributor] (2019). [Google Scholar]
- 21.Zeng Y, Feng Q, Hesketh T, Christensen K, Vaupel JW. Survival, disabilities in activities of daily living, and physical and cognitive functioning among the oldest-old in China: a cohort study. Lancet. (2017) 389:1619–29. doi: 10.1016/s0140-6736(17)30548-2, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Qi J, Zhao N, Liu M, Guo Y, Fu J, Zhang Y, et al. Long-term exposure to fine particulate matter constituents and cognitive impairment among older adults: an 18-year Chinese nationwide cohort study. J Hazard Mater. (2024) 468:133785. doi: 10.1016/j.jhazmat.2024.133785, [DOI] [PubMed] [Google Scholar]
- 23.Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. (2021) 385:1737–49. doi: 10.1056/NEJMoa2102953, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yao X, Han D, Lu Q. The mediating effect of activities of daily living (ADL) disability on the association between depressive symptoms and cognitive impairment in older adults: a national study. BMC Geriatr. (2025) 25:713. doi: 10.1186/s12877-025-06380-z, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gurka MJ, Filipp SL, Musani SK, Sims M, DeBoer MD. Use of BMI as the marker of adiposity in a metabolic syndrome severity score: derivation and validation in predicting long-term disease outcomes. Metabolism. (2018) 83:68–74. doi: 10.1016/j.metabol.2018.01.015, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Li X, Schottker B, Holleczek B, Brenner H. Associations of DNA methylation algorithms of aging and cancer risk: results from a prospective cohort study. EBioMedicine. (2022) 81:104083. doi: 10.1016/j.ebiom.2022.104083, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Huang S, Zhong D, Lv Z, Cheng J, Zou X, Wang T, et al. Associations of multiple plasma metals with the risk of metabolic syndrome: a cross-sectional study in the mid-aged and older population of China. Ecotoxicol Environ Saf. (2022) 231:113183. doi: 10.1016/j.ecoenv.2022.113183, [DOI] [PubMed] [Google Scholar]
- 28.Li X, Schottker B, Holleczek B, Brenner H. Association of longitudinal repeated measurements of frailty index with mortality: cohort study among community-dwelling older adults. EClinicalMedicine. (2022) 53:101630. doi: 10.1016/j.eclinm.2022.101630, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Little R, Rubin D. Multiple Imputation for Nonresponse in Surveys, vol. 10. New York: Wiley; (1987). [Google Scholar]
- 30.Audigier V., White I.R., Jolani S., Debray T.P., Quartagno M., Carpenter J., et al. Multiple Imputation for Multilevel Data with Continuous and Binary Variables. Stat Sci. (2018) 33:160–83, 124. [Google Scholar]
- 31.Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, et al. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA. (2002) 288:2709–16. doi: 10.1001/jama.288.21.2709, [DOI] [PubMed] [Google Scholar]
- 32.Asadi Z, Ghazizadeh H, Timar A, Khodadadegan MA, Ghazavi H, Nasrabadi M, et al. Association between metabolic syndrome and its components with cardiovascular disease risk in the MASHAD cohort study population. J Cardiothorac Med. (2025) 13:1462–75. doi: 10.22038/jctm.2025.70808.1416 [DOI] [Google Scholar]
- 33.Tang X, Wu M, Wu S, Tian Y. Continuous metabolic syndrome severity score and the risk of CVD and all-cause mortality. Eur J Clin Investig. (2022) 52:e13817. doi: 10.1111/eci.13817, [DOI] [PubMed] [Google Scholar]
- 34.Lavie CJ, Milani RV, Ventura HO. Obesity and cardiovascular disease: risk factor, paradox, and impact of weight loss. J Am Coll Cardiol. (2009) 53:1925–32. doi: 10.1016/j.jacc.2008.12.068, [DOI] [PubMed] [Google Scholar]
- 35.Lv Y, Mao C, Gao X, Ji JS, Kraus VB, Yin Z, et al. The obesity paradox is mostly driven by decreased noncardiovascular disease mortality in the oldest old in China: a 20-year prospective cohort study. Nat Aging. (2022) 2:389–96. doi: 10.1038/s43587-022-00201-3, [DOI] [PubMed] [Google Scholar]
- 36.Drame M, Godaert L. The obesity paradox and mortality in older adults: a systematic review. Nutrients. (2023) 15:1780. doi: 10.3390/nu15071780, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Bhaskaran K, Dos-Santos-Silva I, Leon DA, Douglas IJ, Smeeth L. Association of BMI with overall and cause-specific mortality: a population-based cohort study of 3·6 million adults in the UK. Lancet Diabetes Endocrinol. (2018) 6:944–53. doi: 10.1016/s2213-8587(18)30288-2, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Oreopoulos A, Padwal R, Kalantar-Zadeh K, Fonarow GC, Norris CM, McAlister FA. Body mass index and mortality in heart failure: a meta-analysis. Am Heart J. (2008) 156:13–22. doi: 10.1016/j.ahj.2008.02.014, [DOI] [PubMed] [Google Scholar]
- 39.Baumgartner RN, Koehler KM, Gallagher D, Romero L, Heymsfield SB, Ross RR, et al. Epidemiology of sarcopenia among the elderly in New Mexico. Am J Epidemiol. (1998) 147:755–63. doi: 10.1093/oxfordjournals.aje.a009520, [DOI] [PubMed] [Google Scholar]
- 40.Schaap LA, Pluijm SM, Deeg DJ, Visser M. Inflammatory markers and loss of muscle mass (sarcopenia) and strength. Am J Med. (2006) 119:526.e529–17. doi: 10.1016/j.amjmed.2005.10.049 [DOI] [PubMed] [Google Scholar]
- 41.Lindbohm JV, Kaprio J, Korja M. Survival bias explains improved survival in smokers and hypertensive individuals after aSAH. Neurology. (2019) 93:e2105–9. doi: 10.1212/wnl.0000000000008537, [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(A–T) Restricted cubic spline models of the association between MetS Score and all-cause mortality in different predefined subgroups.
Data Availability Statement
Publicly available datasets were analyzed in this study. This data can be found at: https://www.icpsr.umich.edu/web/NACDA/studies/38899.