Abstract
Background
This study aimed to examine the association between the cardiometabolic index (CMI) and intrinsic capacity (IC) and its domains among older adults in China using data from the China Health and Retirement Longitudinal Study (CHARLS).
Methods
Using data from 7,058 participants aged 60–80 years from the 2011–2020 waves of CHARLS, this cross-sectional study examined the association between CMI and IC and its domains. The total IC score was derived from five standardized domains: cognitive, psychological, vitality, locomotion, and sensory. Multiple linear regression models were used to assess the associations between CMI and IC_total as well as its domains. Stratified analyses by sex, age group, and chronic disease status were conducted to examine the robustness of the findings. In addition, an exploratory mediation analysis was performed to assess whether the observed association between CMI and IC was statistically consistent with an indirect association through vitality.
Results
In the crude model, higher CMI was significantly associated with lower IC_total scores (β = -0.049, P < 0.001). In the age- and sex-adjusted model, CMI was positively associated with the cognitive domain (β = 0.084, P < 0.001) and the psychological domain (β = 0.036, P = 0.002), and inversely associated with vitality (β = -0.128, P < 0.001), while no significant association was observed with IC_total. After further adjustment for available sociodemographic and health-related covariates, higher CMI was significantly associated with lower IC_total (β = -0.060, 95% CI: -0.084 to -0.037, P < 0.001) and lower vitality (β = -0.144, 95% CI: -0.167 to -0.121, P < 0.001), whereas the associations with the cognitive, psychological, locomotion, and sensory domains were no longer statistically significant. Mediation analysis suggested that the observed association between CMI and IC was statistically consistent with an indirect association through vitality; however, this finding should be interpreted cautiously given the cross-sectional design.
Conclusions
Among older Chinese adults, higher CMI was independently associated with lower overall intrinsic capacity and lower vitality after adjustment for available covariates. The positive associations with the cognitive and psychological domains observed in simpler models were attenuated after fuller adjustment, suggesting that these associations may be partly explained by confounding. Overall, vitality appears to be the IC domain most consistently associated with cardiometabolic burden in later life.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12877-026-07659-5.
Keywords: Cardiometabolic index, Intrinsic capacity, Older adults, Vitality, CHARLS
Background
In response to the pressing global challenge of population aging, the World Health Organization (WHO) redefined “healthy aging” in 2015 as “the process of developing and maintaining the functional ability that enables well-being in older age”. This framework conceptualizes functional ability as emerging from the dynamic interplay between an individual’s intrinsic capacity (IC), environmental factors, and their reciprocal interactions [1, 2]. IC is defined as the composite of all physical and mental capacities that an individual can draw upon at any point in time. It not only reflects an older adult’s overall health status [3] but also serves as a critical predictor of adverse health outcomes and a foundational metric for advancing healthy aging initiatives [4]. More specifically, IC represents both the physiological reserve accumulated over the life course and a holistic indicator for evaluating the multidimensional health of older adults [5]. It comprises five interrelated domains — cognitive, psychological, vitality, locomotion, and sensory functions — which mutually influence and reinforce one another [6]. Preserving and enhancing IC is therefore pivotal for delaying functional decline, improving quality of life, and fostering resilient, healthy aging trajectories among older populations [7].
The cardiometabolic index (CMI) is a composite biomarker that integrates the triglyceride-to-high-density lipoprotein cholesterol ratio (TG/HDL-C) with the waist-to-height ratio (WHtR), offering a comprehensive assessment of an individual’s risk for cardiometabolic diseases (CMD). This dual-component metric simultaneously captures key pathophysiological features—dyslipidemia, as reflected by an elevated TG/HDL-C ratio, and central adiposity, indexed by WHtR, which serves as a sensitive proxy for visceral fat accumulation [8, 9]. Compared with conventional anthropometric measures such as body mass index (BMI) or waist circumference alone, CMI provides a more nuanced and integrative evaluation of underlying metabolic dysfunction. Moreover, its calculation is non-invasive, straightforward, and readily standardized across diverse clinical and epidemiological settings [10–12]. Emerging evidence further indicates that elevated CMI is independently associated with a higher prevalence of depressive symptoms in middle-aged and older adults, underscoring the potential role of cardiometabolic health optimization in mitigating mood-related disorders [13, 14]. Importantly, heightened cardiometabolic risk—as quantified by CMI—has also been consistently linked to a spectrum of adverse geriatric outcomes, including cognitive frailty, reduced relative handgrip strength, and overall physical frailty [15–18].
Although previous studies have linked CMI to specific health outcomes in older adults, including depressive symptoms, cognitive frailty, and reduced muscle strength, its association with intrinsic capacity as a multidimensional construct remains unclear. Most existing studies have focused on isolated domains or single adverse outcomes rather than examining overall IC and its interrelated components within a unified framework. To address this gap, we used data from the China Health and Retirement Longitudinal Study (CHARLS) to investigate the association between CMI and overall IC as well as its five domains among older Chinese adults. We hypothesized that higher CMI would be associated with lower IC, with potentially differential associations across domains. By clarifying these relationships, this study aims to provide evidence for understanding how cardiometabolic burden is associated with functional reserve and healthy ageing in later life.
Materials and methods
CHARLS
The China Health and Retirement Longitudinal Study (CHARLS) is a nationally representative, longitudinal survey targeting community-dwelling adults aged 45 years and older in China [19]. Launched in 2011, CHARLS employs a complex multistage probability sampling design that encompasses 150 county-level units and 450 communities (including villages) across 28 provinces, autonomous regions, and municipalities directly under the central government, thereby ensuring national representativeness. The baseline wave enrolled 19,000 individuals from 12,400 households. Subsequent biennial follow-up surveys were conducted in 2013, 2015, 2018, and 2020, utilizing face-to-face, computer-assisted personal interviews. The questionnaire captures comprehensive multidimensional data on participants’ sociodemographic characteristics, household composition, health status and functional capacity, healthcare utilization and insurance coverage, employment, retirement, pension arrangements, income, expenditures, asset holdings, and housing conditions. In addition, objective biomarker data—including standardized anthropometric measurements (e.g., waist circumference and height) and blood-based biochemical assays—are systematically collected. The study strictly adheres to established ethical standards: all participants provided written informed consent, and the protocol was approved by the Biomedical Ethics Review Committee of Peking University (Approval No. IRB00001052-11015). Further methodological details have been published elsewhere [20].
Study sample
In this study, we analyzed data from 7,058 participants aged 60–80 years drawn from the CHARLS database covering the period 2011–2020. The study selection process and inclusion criteria are summarized in Fig. 1. Participants were excluded if they met any of the following conditions: (1) missing IC assessments (n = 2,260); (2) missing CMI values (n = 7,975); (3) missing demographic or covariate information (n = 79,335). After applying these exclusion criteria, a final analytic sample of 7,058 individuals was retained.
Fig. 1.
Flow diagram of the screening of the CHARLS participants. CHARLS, China Health and Retirement Longitudinal Study; BMI, body mass index; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; CRP, C-reactive protein; ADL, basic activities of daily living
Intrinsic capacity
This study operationalized intrinsic capacity (IC) according to the multidomain framework proposed in the WHO Integrated Care for Older People (ICOPE) approach, while adapting the measurement to the variables available in CHARLS [21]. Because CHARLS is a secondary dataset and does not include the full set of standardized ICOPE instruments, several IC domains were constructed using proxy indicators based on conceptual relevance and variable availability. Specifically, (1) Cognitive: Cognitive function was assessed through the Telephone Interview for Cognitive Status (TICS), adapted from the Health and Retirement Study (HRS) conducted in the USA [22, 23]. In our evaluation, cognitive function encompasses intelligence level and situational memory as per the CHARLS assessment tools, with higher scores indicating superior cognitive performance. (2) Psychological [24]: This dimension combined reverse-scored CESD-10 and life satisfaction to capture complementary aspects of psychological functioning, namely emotional distress and subjective well-being. (3) Vitality: Vitality was defined using maximum grip strength and inverse log-transformed C-reactive protein (CRP) to reflect both physical reserve and inflammatory status. (4) Locomotion: Locomotion function was measured by walking speed over a distance of 2.5 m, with an acceptable range defined as 0.2 to 2.0 m per second. Data points falling outside this range were excluded from analysis. (5) Sensory: Sensory function was based on self-reported vision and hearing because objective sensory testing was not available in CHARLS. All components were standardized before aggregation so that the composite scores reflected relative performance within each domain.
Cardiometabolic index
CMI is derived from four key anthropometric and biochemical variables: triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), waist circumference, and height. The index is calculated using the following formula: CMI = TG (mmol/L) /HDL-C (mmol/L) × waist circumference (cm)/height (cm) [25, 26]. In the CHARLS cohort, venous blood samples were collected by trained medical personnel from participants who provided written informed consent [27, 28]. Detailed information regarding the study design, data collection protocols, and methodological procedures is publicly available on the official CHARLS website (https://charls.pku.edu.cn/wd/a2011nqgjxdc.htm).
CMI was calculated as the product of the triglyceride-to-high-density lipoprotein cholesterol ratio (TG/HDL-C) and the waist-to-height ratio (WHtR). To ensure consistency in the lipid ratio, TG and HDL-C were expressed in the same unit before CMI calculation. In the present study, laboratory values were converted to mmol/L prior to CMI calculation. For descriptive purposes, TG and HDL-C are presented in Table 1 in mg/dL, consistent with the original laboratory reporting format of CHARLS.
Table 1.
Baseline characteristics of the CHARLS participants by cardiometabolic index quartile
| Characteristics | Total N = 7058 |
Quartile 1 N = 1765 |
Quartile 2 N = 1764 |
Quartile 3 N = 1765 |
Quartile 4 N = 1764 |
P value |
|---|---|---|---|---|---|---|
| Age, mean(SD), years | 66.79(5.17) | 67.16(5.29) | 66.99(5.16) | 66.64(5.20) | 66.39(5.01) | < 0.001 |
| Sex, n (%) | < 0.001 | |||||
| Female | 3143(44.50) | 574(32.50) | 714(40.50) | 889(50.40) | 966(54.80) | |
| Male | 3915(55.50) | 1191(67.50) | 1050(59.50) | 876(49.60) | 798(45.20) | |
| Marital status | 0.076 | |||||
| Married | 5921(83.90) | 1488(84.30) | 1449(82.10) | 1479(83.80) | 1505(85.30) | |
| Others | 1136(16.10) | 277(15.70) | 315(17.90) | 285(16.20) | 259(14.7) | |
| Residence, n (%) | < 0.001 | |||||
| Rural | 2714(38.50) | 502(28.40) | 619(35.10) | 755(42.80) | 838(47.50) | |
| Urban | 4344(61.50) | 1263(71.60) | 1145(64.90) | 1010(57.20) | 926(52.50) | |
| Household registration, n (%) | < 0.001 | |||||
| Agricultural | 1583(22.40) | 263(15.70) | 346(20.30) | 474(28.00) | 500(29.70) | |
| Non-agricultural | 5172(73.30) | 1414(84.30) | 1355(79.70) | 1221(72.00) | 1182(70.30) | |
| Smoking status, n (%) | < 0.001 | |||||
| No | 3575(50.70) | 750(42.50) | 844(47.80) | 978(55.40) | 1003(56.90) | |
| Yes | 3483(49.30) | 1015(57.50) | 920(52.20) | 787(44.60) | 761(43.10) | |
| Drinking status, n (%) | < 0.001 | |||||
| No | 4572(64.80) | 942(53.40) | 1149(65.10) | 1223(69.30) | 1258(71.30) | |
| Yes | 2486(35.20) | 823(46.60) | 615(34.90) | 542(30.70) | 506(28.70) | |
| Chronic disease status, n (%) | < 0.001 | |||||
| No | 1359(19.30) | 425(24.10) | 347(19.70) | 328(18.60) | 259(14.70) | |
| Yes | 5699(80.70) | 1340(75.90) | 1417(80.30) | 1437(81.40) | 1505(85.30) | |
| Height, mean(SD), m | 1.58(0.08) | 1.58(0.08) | 1.58(0.08) | 1.58(0.08) | 1.58(0.08) | 0.247 |
| Waist circumference, mean(SD), cm | 85.05(13.10) | 75.17(15.33) | 83.75(10.11) | 88.54(9.76) | 92.73(9.15) | < 0.001 |
| BMI, mean(SD), kg/m2 | 23.42(3.69) | 21.04(2.85) | 22.80(3.42) | 24.26(3.38) | 25.57(3.43) | < 0.001 |
| TG, mean(SD), mg/dL | 134.05(92.65) | 68.67(20.38) | 94.79(22.48) | 130.63(30.92) | 242.15(122.31) | < 0.001 |
| HDL-C, mean(SD), mg/dL | 51.38(13.48) | 64.48(13.51) | 52.80(9.97) | 47.74(8.49) | 40.50(8.27) | < 0.001 |
| Cognitive function | 11.20(3.60) | 10.92(3.58) | 11.12(3.61) | 11.21(3.62) | 11.55(3.56) | < 0.001 |
| Situational memory | 3.21(1.67) | 3.05(1.66) | 3.17(1.65) | 3.24(1.63) | 3.37(1.71) | < 0.001 |
| Intelligence level | 7.99(2.63) | 7.87(2.64) | 7.94(2.68) | 7.98(2.67) | 8.18(2.53) | 0.004 |
| Mental health | 8.15(6.26) | 8.43(6.27) | 8.03(6.31) | 8.11(6.25) | 8.03(6.22) | 0.183 |
| Life satisfaction | 3.30(0.74) | 3.28(0.73) | 3.30(0.74) | 3.29(0.76) | 3.31(0.75) | 0.656 |
| Handgrip strength (left) | 28.18(9.54) | 28.76(9.09) | 28.78(9.32) | 27.67(9.73) | 27.51(9.94) | < 0.001 |
| Handgrip strength (right) | 29.51(9.94) | 30.15(9.56) | 29.68(9.87) | 29.12(10.04) | 29.11(10.25) | 0.004 |
| CRP, mean(SD), mg/L | 2.94(6.96) | 2.65(7.88) | 2.58(6.66) | 3.10(7.91) | 3.43(4.95) | 0.001 |
| Walking speed | 3.80(1.91) | 3.75(1.55) | 3.72(1.84) | 3.82(1.76) | 3.90(2.37) | 0.030 |
| Distance vision, n (%) | 0.838 | |||||
| 1 | 1725(24.40) | 422(23.90) | 438(24.80) | 423(24.00) | 442(25.10) | |
| 2 | 3336(47.30) | 829(47.00) | 833(47.20) | 859(48.70) | 815(46.20) | |
| 3 | 1139(16.10) | 304(17.20) | 275(15.60) | 268(15.20) | 292(16.60) | |
| 4 | 732(10.40) | 175(9.90) | 188(10.70) | 189(10.70) | 180(10.20) | |
| 5 | 126(1.80) | 35(2.00) | 30(1.70) | 26(1.50) | 35(2.00) | |
| Near vision, n (%) | 0.255 | |||||
| 1 | 1422(20.10) | 332(18.80) | 370(21.00) | 347(19.70) | 373(21.10) | |
| 2 | 3522(49.90) | 907(51.40) | 855(48.50) | 889(50.40) | 871(49.40) | |
| 3 | 1360(19.30) | 345(19.50) | 329(18.70) | 346(19.60) | 340(19.30) | |
| 4 | 645(9.10) | 153(8.70) | 187(10.60) | 148(8.40) | 157(8.90) | |
| 5 | 109(1.50) | 28(1.60) | 23(1.30) | 35(2.00) | 23(1.30) | |
| Auditory, n (%) | 0.265 | |||||
| 1 | 1071(15.20) | 277(15.70) | 295(16.70) | 258(14.60) | 241(13.70) | |
| 2 | 3672(52.00) | 931(52.70) | 898(50.90) | 926(52.50) | 917(52.00) | |
| 3 | 1394(19.80) | 337(19.10) | 329(18.70) | 362(20.50) | 366(20.70) | |
| 4 | 829(11.70) | 193(10.90) | 223(12.60) | 199(11.30) | 214(12.10) | |
| 5 | 92(1.30) | 27(1.50) | 19(1.10) | 20(1.10) | 26(1.50) | |
| ADL, n (%) | 0.601 | |||||
| Independent | 5563(78.80) | 1412(80.00) | 1421(80.60) | 1377(78.00) | 1353(76.70) | |
| Not independent | 1495(21.20) | 353(20.00) | 343(19.40) | 388(22.00) | 411(23.30) | |
| IADL, n (%) | 0.006 | |||||
| Independent | 5501(77.90) | 1400(79.30) | 1383(78.40) | 1354(76.70) | 1364(77.30) | |
| Not independent | 1557(22.10) | 365(20.70) | 381(21.60) | 411(23.30) | 400(22.70) |
Abbreviations: CHARLS China Health and Retirement Longitudinal Study, BMI body mass index, TG triglycerides, HDL-C high-density lipoprotein cholesterol, CRP C-reactive protein, ADL activities of daily living, IADL instrumental activities of daily living, SD standard deviation. TG and HDL-C are presented in mg/dL in Table 1 for descriptive purposes; values were converted to mmol/L before CMI calculation
Covariates
Covariates related to IC and CMI were also included in this study for analysis: age, sex (male, female), marital status (married, other), place of residence (rural, urban), household registration (agricultural, non-agricultural), smoking (yes, no), alcohol consumption (yes, no), chronic disease status (yes, no), BMI, ADLs (dressing, bathing, eating, getting in and out of bed, using the toilet, and controlling urination and defecation), and IADLs (household chores, cooking, shopping, managing money, and taking medications).
Statistical analysis
Participants were stratified into four groups (Q1–Q4) based on CMI quartiles, and baseline characteristics were described for each group. Continuous variables are presented as means and standard deviations (mean ± SD), and categorical variables as frequencies and percentages (n, %). Group differences in continuous variables were assessed using one-way analysis of variance (ANOVA), and differences in categorical variables were assessed using the chi-square test. Ordered categorical variables were compared using the Wilcoxon rank-sum test.
To investigate the association between CMI and IC and its domains, we constructed sequential multiple linear regression models. Model 0 was unadjusted. Model 1 was adjusted for age and sex. Model 2 was further adjusted for marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL. Separate regression models were fitted for IC_total and for each IC domain. Results are presented as standardized regression coefficients (β), together with 95% confidence intervals (CIs) and corresponding P values.
To evaluate the robustness of the association between CMI and the vitality domain and to examine potential heterogeneity across key subpopulations, we conducted stratified analyses by sex, age group (60–69 years vs. 70–80 years), and chronic disease status. Within each stratum, linear regression models were fitted with adjustment for the remaining covariates.
Furthermore, to explore the relationship between CMI and IC_total, we performed an exploratory mediation analysis to assess whether the observed association between CMI and IC was statistically consistent with an indirect association through the vitality domain. Using the bootstrap method with 5,000 repetitions, we estimated the average causal mediation effect (ACME), average direct effect (ADE), and total effect, together with their 95% confidence intervals. The mediation model was adjusted for the same full set of covariates included in Model 2, namely age, sex, marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL. Given the cross-sectional design of the present study, this analysis was intended to evaluate statistical consistency with a hypothesized relational structure rather than to establish causal pathways.
To assess the robustness of the domain operationalization, we conducted sensitivity analyses by reconstructing the vitality domain using grip strength alone and by analyzing reverse-scored CESD-10 and life satisfaction separately instead of combining them into a single psychological domain score.
Results
Characteristics of the participants
Table 1 presents the baseline characteristics of participants stratified by CMI quartiles. The mean age of participants was 66.79 years (SD, 5.17), with 3,143 participants (44.50%) being female. Participants in higher quartiles were more likely to be female, rural residents, with agricultural household registration, chronic diseases, and higher waist circumference and BMI levels, Compared to those in the low quartile, they also had higher TG levels and lower HDL levels. Additionally, they exhibited higher cognitive ability scores, higher episodic memory scores, and elevated CRP levels.
Association between CMI and IC and its domains
As shown in Table 2, in the crude model (Model 0), higher CMI was significantly associated with lower IC_total scores (β = -0.049, P < 0.001). After adjustment for age and sex (Model 1), the association between CMI and IC_total was attenuated and no longer statistically significant (β = -0.011, P = 0.344). In the age- and sex-adjusted model, higher CMI was positively associated with the cognitive domain (β = 0.084, P < 0.001) and the psychological domain (β = 0.036, P = 0.002), and inversely associated with vitality (β = -0.128, P < 0.001), while no significant associations were observed for locomotion or sensory function.
Table 2.
Association between CMI and IC and its domains
| Crude model (Model 0) | Adjusted model (Model 1) | Adjusted model (Model 2) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β | SE | 95% CI | P-value | β | SE | 95% CI | P-value | β | SE | 95% CI | P-value | |
| IC_total | -0.049 | 0.012 | -0.073, -0.026 | < 0.001 | -0.011 | 0.011 | -0.033, 0.011 | 0.344 | -0.060 | 0.012 | -0.084, -0.037 | < 0.001 |
| Cognitive | 0.062 | 0.012 | 0.039, 0.085 | < 0.001 | 0.084 | 0.012 | 0.061, 0.107 | < 0.001 | 0.022 | 0.012 | -0.002, 0.046 | 0.068 |
| Psychological | 0.018 | 0.012 | -0.006, 0.041 | 0.137 | 0.036 | 0.012 | 0.013, 0.060 | 0.002 | -0.001 | 0.013 | -0.025, 0.024 | 0.961 |
| Vitality | -0.181 | 0.012 | -0.204, -0.158 | < 0.001 | -0.128 | 0.011 | -0.148, -0.107 | < 0.001 | -0.144 | 0.012 | -0.167, -0.121 | < 0.001 |
| Locomotion | -0.013 | 0.012 | -0.037, 0.010 | 0.263 | -0.005 | 0.012 | -0.029, 0.018 | 0.644 | -0.017 | 0.013 | -0.043, 0.009 | 0.194 |
| Sensory | -0.006 | 0.012 | -0.029, 0.017 | 0.610 | -0.014 | 0.012 | -0.037, 0.010 | 0.254 | -0.007 | 0.013 | -0.033, 0.019 | 0.603 |
Abbreviations: β standardized regression coefficient, SE standard error, CI confidence interval, CMI cardiometabolic index, IC intrinsic capacity, ADL activities of daily living, IADL instrumental activities of daily living
Model 0: unadjusted model
Model 1: adjusted for age and sex
Model 2: further adjusted for marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL
After further adjustment for marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL (Model 2), the pattern of associations changed. Higher CMI was significantly associated with lower IC_total (β = -0.060, 95% CI: -0.084 to -0.037, P < 0.001) and lower vitality (β = -0.144, 95% CI: -0.167 to -0.121, P < 0.001). In contrast, the associations with the cognitive domain (β = 0.022, P = 0.068) and the psychological domain (β = -0.001, P = 0.961) were attenuated and were no longer statistically significant. No significant associations were observed for locomotion or sensory function in Model 2.
Overall, these findings indicate that the association between CMI and IC is sensitive to the degree of covariate adjustment. Among the five domains, the composite vitality domain showed the most consistent inverse association with CMI, and a significant adverse association with overall IC emerged after more comprehensive adjustment.
Moreover, to visually depict this heterogeneous association pattern, Fig. 2 demonstrates that, under conditions of matching age, sex, marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL, individuals with higher CMI exhibit lower vitality function. This dose–response relationship is further quantified in trend plots based on CMI quartiles. As the CMI quartile increases from the lowest (Q1) to the highest quartile (Q4), mean vitality scores decline monotonically. Figure 3 further illustrates this monotonic decline across increasing CMI quartiles.
Fig. 2.
Association between standardized CMI and vitality in older Chinese adults (n = 7,058) after adjustment for age, sex, marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL. The solid line represents the fitted association between CMI and vitality, and the shaded area indicates the 95% confidence interval. Both variables were standardized (mean = 0, SD = 1). Darker areas indicate a higher density of overlapping observations
Fig. 3.
Dose–response relationship between CMI quartiles and vitality in older Chinese adults (n = 7,058) based on CHARLS data. Mean vitality z-scores decreased progressively from the lowest CMI quartile (Q1) to the highest quartile (Q4). Error bars represent 95% confidence intervals
Sensitivity analyses
Sensitivity analyses showed that the findings differed according to the operationalization of specific IC domains. When vitality was represented by grip strength alone, the association with CMI was positive in Model 1 (β = 0.045, 95% CI: 0.027 to 0.063, P < 0.001) but became non-significant in Model 2 (β = -0.010, 95% CI: -0.030 to 0.009, P = 0.312). This indicates that the inverse association observed for the composite vitality domain was not explained by grip strength alone. When the psychological domain was decomposed into its two components, reverse-scored CESD-10 was positively associated with CMI in Model 1 but not in Model 2 (β = 0.004, 95% CI: -0.020 to 0.028, P = 0.737), while life satisfaction was not significantly associated with CMI in either model (Model 2: β = -0.005, 95% CI: -0.031 to 0.021, P = 0.706). These results suggest that the null association for the psychological domain after fuller adjustment was robust to alternative domain definitions (Supplementary Table S1).
Subgroup analyses
Subgroup analyses showed that the inverse association between CMI and vitality remained statistically significant across all examined strata (Table 3). Significant inverse associations were observed in both females (β = -0.153, P < 0.001) and males (β = -0.148, P < 0.001), in participants aged 60–69 years (β = -0.142, P < 0.001) and 70–80 years (β = -0.160, P < 0.001), and in those with and without chronic disease. Formal interaction tests further showed a statistically significant interaction by sex (P < 0.001), a modest interaction by age group (P = 0.045), and no significant interaction by chronic disease status (P = 0.520).
Table 3.
Subgroup analyses of the association between CMI and vitality, with formal interaction tests
| Characteristics | n | β | 95% CI | P value | P for interaction |
|---|---|---|---|---|---|
| Sex | < 0.001 | ||||
| Female | 3143 | -0.153 | (-0.183, -0.123) | < 0.001 | |
| Male | 3915 | -0.148 | (-0.182, -0.115) | < 0.001 | |
| Age | 0.045 | ||||
| 60–69 | 5025 | -0.142 | (-0.168, -0.116) | < 0.001 | |
| 70–80 | 2033 | -0.160 | (-0.210, -0.110) | < 0.001 | |
| Chronic disease status | 0.520 | ||||
| no | 1359 | -0.148 | (-0.201, -0.096) | < 0.001 | |
| yes | 5699 | -0.143 | (-0.168, -0.117) | < 0.001 |
Abbreviations: β standardized regression coefficient, β for the association between CMI and vitality, CI confidence interval, CMI cardiometabolic index. All subgroup models were adjusted for age, sex, marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL, where applicable. P for interaction was obtained from the fully adjusted model by including the corresponding interaction term between CMI and the subgroup variable
Mediation analysis
Mediation analysis suggested that the association between CMI and IC_total was statistically consistent with an indirect association through vitality. Specifically, a negative indirect association and a positive direct association were observed, indicating that the two components operated in opposite directions, and the total association remained statistically significant (Fig. 4; Table 4). These findings suggest that vitality may statistically account for part of the observed association between CMI and overall IC. However, because this analysis was based on cross-sectional data, these results should be interpreted as exploratory and associational rather than causal.
Fig. 4.
Decomposition of total, direct, and indirect associations in the exploratory mediation analysis. Abbreviations: ACME, indirect association estimated in the mediation model; ADE, direct association estimated in the mediation model; SE, standard error; LLCI, lower limit of the 95% confidence interval; ULCI, upper limit of the 95% confidence interval. ***p < 0.001
Table 4.
Decomposition of total, direct, and indirect associations in the exploratory mediation analysis
| Effect | SE | LLCI | ULCI | P value | |
|---|---|---|---|---|---|
| Total | -0.0603 | 0.0119 | -0.0837 | -0.0369 | < 0.001 |
| ADE | 0.0100 | 0.0106 | -0.0107 | 0.0308 | 0.3430 |
| ACME | -0.0703 | 0.0066 | -0.0836 | -0.0572 | - |
Abbreviations: ACME indirect association estimated in the mediation model, ADE direct association estimated in the mediation model, SE standard error, LLCI lower limit of the 95% confidence interval, ULCI upper limit of the 95% confidence interval. The mediation model was adjusted for age, sex, marital status, place of residence, household registration, smoking status, alcohol consumption, chronic disease status, BMI, ADL, and IADL
Discussion
In this nationally representative sample of older Chinese adults, we found that the association between CMI and IC depended on the degree of covariate adjustment. In the age- and sex-adjusted model, higher CMI was positively associated with the cognitive and psychological domains and negatively associated with vitality, whereas no significant association was observed with overall IC. However, after further adjustment for sociodemographic and health-related covariates, the positive associations with the cognitive and psychological domains were attenuated and were no longer statistically significant. In contrast, higher CMI remained significantly associated with lower overall IC and lower vitality. Taken together, these findings suggest that the most robust association between CMI and IC in later life is concentrated in the composite vitality domain, with an additional adverse association with global IC.
The inverse association between CMI and the composite vitality domain was biologically plausible and remained evident in the main analyses. In the present study, vitality was represented by handgrip strength and inflammatory status, both of which reflect physiological reserve and metabolic health [29, 30]. Previous studies have shown that greater muscle strength is associated with lower cardiometabolic risk [16, 31], whereas metabolic abnormalities are linked to poorer muscle function [32–35]. A higher CMI reflects central adiposity and an adverse lipid profile, both of which are closely related to chronic low-grade inflammation, insulin resistance, and impaired skeletal muscle metabolism [36, 37]. These mechanisms may accelerate muscle catabolism, reduce physiological reserve, and contribute to lower vitality in older adults. However, the sensitivity analyses suggested that this finding was partly dependent on domain operationalization, as the association was no longer statistically significant when vitality was represented by grip strength alone after full adjustment. This pattern suggests that the composite vitality score may capture a broader construct integrating muscular strength and inflammatory status, rather than reflecting handgrip strength alone.
We also observed evidence of subgroup heterogeneity in the association between CMI and vitality. Formal interaction testing showed a significant interaction by sex and a modest interaction by age group, whereas no significant interaction was observed for chronic disease status. A possible explanation for the stronger association in women is that females, particularly after menopause, may experience changes in fat distribution, muscle mass, and estrogen-related metabolic regulation that modify the relationship between cardiometabolic burden and vitality [38, 39]. In addition, sex differences in body composition and inflammatory response may alter how cardiometabolic burden is translated into vitality-related impairment [40, 41]. By contrast, although the interaction by age group reached borderline statistical significance, the subgroup-specific estimates were relatively similar, and this finding should therefore be interpreted cautiously. Overall, these results suggest that sex may be a more relevant effect modifier than chronic disease status in this dataset in the association between CMI and vitality, although further confirmation is needed in longitudinal studies.
By contrast, the positive associations of CMI with the cognitive and psychological domains in the simpler model did not persist after fuller adjustment. This suggests that these initial associations may have been partly explained by residual confounding related to social background, health status, or functional dependence [42]. Reverse causality is also possible in a cross-sectional setting, as individuals with better preserved cognition or psychological functioning may differ systematically in nutritional status, health behaviors, or disease burden. It is also possible that very low cardiometabolic status in older adults may coexist with malnutrition, frailty, or reduced physiological reserve, which are themselves associated with poorer cognitive and psychological performance [43–47]. In addition, previous studies linking CMI to depressive symptoms have generally focused on adverse mental health outcomes rather than positive psychological function per se [26, 48]. Moreover, even in the age- and sex-adjusted model, the positive association with cognition was modest in magnitude and should be interpreted cautiously in terms of potential clinical relevance. Therefore, the present findings do not support an independent beneficial association of higher CMI with cognitive or psychological function in older adults.
The sensitivity analyses further informed the interpretation of the domain operationalization. When vitality was represented by grip strength alone, the inverse association with CMI observed for the composite vitality domain was no longer evident after full adjustment. By contrast, the results for the psychological domain were more stable in interpretation: after fuller adjustment, neither reverse-scored CESD-10 nor life satisfaction was significantly associated with CMI, supporting our conclusion that the psychological domain was not independently associated with CMI in the fully adjusted model. Because the sensory domain relied on self-reported vision and hearing, it should be interpreted as a proxy measure and may be subject to reporting bias. More broadly, these findings highlight that the operationalization of IC domains in secondary datasets such as CHARLS is necessarily constrained by variable availability, and that alternative domain definitions may influence the strength of the observed associations [24, 49, 50].
No significant independent associations were observed between CMI and the locomotion or sensory domains after extended adjustment. This may reflect the fact that these domains are influenced more strongly by long-term structural ageing processes than by relatively proximal metabolic disturbance [51, 52]. Walking performance and sensory impairment often result from cumulative neurodegenerative, musculoskeletal, or organ-specific changes that may not be fully captured by CMI. In addition, the measures used in this study may be less sensitive to short- to medium-term metabolic effects than the vitality domain [53]. This may explain why the associations for locomotion and sensory function were weaker and less consistent.
The exploratory mediation analysis suggested that the association between CMI and IC_total was statistically consistent with an indirect association through vitality. Importantly, the mediation model was adjusted for the same full set of covariates as the main regression model, which improves the comparability of the mediation findings with the fully adjusted associations reported in the primary analyses. The indirect and direct associations were observed in opposite directions, a pattern that is conceptually related to statistical suppression. However, because the total association remained statistically significant in the analysis, this finding is more appropriately described as opposing direct and indirect associations rather than as a classic suppression pattern leading to a null total association. In the setting of cross-sectional data, mediation analysis does not establish temporal order or causal pathways, but only evaluates whether the observed data are statistically consistent with a hypothesized relational structure [54]. Therefore, rather than inferring that vitality causally mediates the relationship between CMI and IC, our findings are more appropriately interpreted as indicating that vitality may represent an important dimension through which CMI is associated with overall IC.
From a clinical and public health perspective, these findings suggest that CMI may be more useful as a marker of vulnerability to reduced physiological reserve and lower IC than as an indicator of advantage in selected functional domains. In particular, the consistent association with vitality highlights the relevance of cardiometabolic health to muscle strength, inflammatory burden, and functional resilience in later life. These findings also suggest that preserving cardiometabolic health in older adults may have value not only for reducing cardiovascular and metabolic risk, but also for maintaining vitality and overall functional reserve as central components of healthy ageing.
Several limitations should be considered. First, the cross-sectional design precludes causal inference, and reverse causation cannot be excluded. Second, some potentially important confounders, including education, household economic status, physical activity, diet, sleep, and detailed medication use, were not available in the analytic dataset and could not be included in the models. Third, the IC domains were constructed from proxy measures available in CHARLS and may not fully capture the original WHO framework. In addition, the sensitivity analyses suggested that some findings, particularly for vitality, were sensitive to domain definition, and the sensory domain may also be affected by reporting bias because it relied on self-reported vision and hearing. Finally, selection bias related to missing data cannot be completely excluded. Longitudinal studies with richer socioeconomic, behavioral, and biological information are needed to clarify the temporal and mechanistic relationships between cardiometabolic burden and trajectories of IC in older adults.
Conclusions
In conclusion, higher CMI was independently associated with lower overall IC and lower vitality among older Chinese adults after adjustment for available covariates. The positive associations with the cognitive and psychological domains observed in simpler models were not sustained after fuller adjustment. These findings suggest that the composite vitality domain may be the IC component most consistently associated with cardiometabolic burden in later life and highlight the importance of cardiometabolic health in the preservation of functional reserve during ageing.
Supplementary Information
Authors’ contributions
ZC and YL conceived and designed the study. ZC performed the statistical analyses and drafted the manuscript. HW, JZ, GG, SW, SZ, JD, LS, and SZ contributed to data interpretation and manuscript revision. YL supervised the study and critically revised the manuscript. All authors read and approved the final manuscript.
Funding
This study was supported by the Postgraduate Research & Practice Innovation Program of Jiangsu Province (No. SJCX25_2410) and the National Social Science Foundation of China 2024 (Grant Number 24BRK008).
Data availability
The datasets used and/or analyzed during the current study are available from the China Health and Retirement Longitudinal Study (CHARLS) repository. CHARLS data are publicly accessible upon reasonable request/registration through the official website: https://charls.pku.edu.cn/.
Declarations
Ethics approval and consent to participate
The CHARLS was conducted with ethical approval from the Biomedical Ethics Committee of Peking University (IRB00001052-11015). Written informed consent was obtained from all participants during the original CHARLS data collection process. This study only used anonymized aggregated data without access to any personal identifiers, thus complying with ethical standards. No additional ethical approval was required for this secondary data analysis.
Consent for publication
Not Applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Ziyue Chen is the first author.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.World Health Organization. World report on aging and health[R/OL]. (2015-09-29). https://www.afro.who.int/publications/worldreport-ageing-and-health.
- 2.Venkatapuram S, Amuthavalli Thiyagarajan J. The capability approach and the WHO healthy ageing framework (for the UN Decade of Healthy Ageing). Age Ageing. 2023;52(Suppl 4):iv6–9. 10.1093/ageing/afad126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.World Health Organization. World report on ageing and health[EB/OL]. [2024-01-04]. https://www.who.int/publications/i/item/9789241565042.
- 4.Blancafort AS, Cuevas-Lara C, Martínez-Velilla N, Zambom-Ferraresi F, Soto ME, Tavassoli N, et al. A multi- domain group-based intervention to promote physical activity, healthy nutrition, and psychological wellbeing in older people with losses in intrinsic capacity: AMICOPE Development Study. Int J Environ Res Public Health. 2021;18(11):5979. 10.3390/ijerph18115979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Liang Y, Shang S, Gao Y, Zhai J, Cheng X, Yang C, et al. Measurements of Intrinsic Capacity in Older Adults: A Scoping Review and Quality Assessment. J Am Med Dir Assoc. 2023;24(3):267–e2762. 10.1016/j.jamda.2022.09.011. [DOI] [PubMed] [Google Scholar]
- 6.Gutiérrez-Robledo LM, García-Chanes RE, Pérez-Zepeda MU. Allostatic Load as a Biological Substrate to Intrinsic Capacity: A Secondary Analysis of CRELES. J Nutr Health Aging. 2019;23(9):788–95. 10.1007/s12603-019-1251-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.World Health Organization. Integrated care for older people (ICOPE): guidance for person-centred assessment and pathways in primary care. 2019. (Accessed 10 Apr 2024). https://www.who.int/publications/i/item/WHO-FWC-ALC-19.1.
- 8.Qiu Y, Yi Q, Li S, Sun W, Ren Z, Shen Y, et al. Transition of cardiometabolic status and the risk of type 2 diabetes mellitus among middle-aged and older Chinese: A National cohort study. J Diabetes Invest. 2022;13(8):1426–37. 10.1111/jdi.13805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Xu B, Wu Q, Yin G, Lu L, La R, Zhang Y, et al. Associations of cardiometabolic index with diabetic statuses and insulin resistance: the mediating role of inflammation-related indicators . BMC Public Health. 2024;24(1):2736. 10.1186/s12889-024-20048-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wang H, Chen Y, Sun G, Jia P, Qian H, Sun Y. Validity of cardiometabolic index, lipid accumulation product, and body adiposity index in predicting the risk of hypertension in Chinese population. Postgrad Med. 2018;130(3):325–33. 10.1080/00325481.2018.1444901. [DOI] [PubMed] [Google Scholar]
- 11.Wei J, Liu X, Xue H, Wang Y, Shi Z. Comparisons of visceral adiposity index, body shape index, bodymass index and waist circumference and their associations with diabetes mellitus in adults. Nutrients. 2019;11:1580. 10.3390/nu11071580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Amato MC, Guarnotta V, Giordano C. Body composition assessment for the definition of cardiometabolic risk. J Endocrinol Invest. 2013;36:537–43. 10.3275/8943. [DOI] [PubMed] [Google Scholar]
- 13.Han S, Zhang Y, Wu B, Chen Q, Han Z, Chen J, et al. Curvilinear association between cardiometabolic index and depressive symptoms in individuals aged 45 and older: a cross-sectional study of CHARLS. Front Public Health. 2025;13:1534302. 10.3389/fpubh.2025.1534302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cheng L, Wu Q, Wang S. Cardiometabolic index is associated with Increased depression: a population-based study. J Affect Disord. 2024;348:259–64. 10.1016/j.jad.2023.12.073. [DOI] [PubMed] [Google Scholar]
- 15.Yan H, Lang J, Li C, Eftekhariranjbar S, Jiang G, Lei J, et al. Cognitive frailty and cardiometabolic risk in middle-aged and older adults: evidence from the UK and China. Aging Clin Exp Res. 2025;37(1):269. 10.1007/s40520-025-03179-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hao G, Chen H, Ying Y, Wu M, Yang G, Jing C. The Relative Handgrip Strength and Risk of Cardiometabolic Disorders: A Prospective Study. Front Physiol. 2020;11:719. 10.3389/fphys.2020.00719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhou K, Wang A, Yi K. Cardiometabolic multimorbidity and frailty in middle-aged and older adults: a cross-nationally harmonized study. Front Public Health. 2025;13:1565682. 10.3389/fpubh.2025.1565682. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang W, Wang K, Chen G, Niu J, Liu Y, Wang Z, et al. Changes in Frailty and Cardiometabolic Disease: Evidence From a Longitudinal Cohort Study and Genetic Analysis. JACC Adv. 2025;4(6 Pt 2):101782. 10.1016/j.jacadv.2025.101782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhao B, Liu Z, Fu Y, Zhang H, Wu J, Lai C, et al. Social determinants of intrinsic capacity: a national cohort study. Am J Prev Med. 2024;66(3):559–67. 10.1016/j.amepre.2023.10.008. [DOI] [PubMed] [Google Scholar]
- 20.Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China health and retirement longitudinal study (CHARLS). Int J Epidemiol. 2014;43:61–8. 10.1093/ije/dys203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Beard JR, Si Y, Liu Z, Chenoweth L, Hanewald K. Intrinsic Capacity: Validation of a New WHO Concept for Healthy Aging in a Longitudinal Chinese Study. J Gerontol Biol Sci Med Sci. 2022;77:94–100. 10.1093/gerona/glab226. [DOI] [PubMed] [Google Scholar]
- 22.Feng Q, Yang S, Ye S, Wan C, Wang H, You J. Mediation of depressive symptoms in the association between blood urea nitrogen to creatinine ratio and cognition among middle-aged and elderly adults: evidence from a national longitudinal cohort study. BMC Psychiatry. 2024;24(1):515. 10.1186/s12888-024-05941-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mo S, Wang Y, Peng M, Wang Q, Zheng H, Zhan Y. Sex disparity in cognitive aging related to later-life exposure to ambient air pollution. Sci Total Environ. 2023;886:163980. 10.1016/j.scitotenv.2023.163980. [DOI] [PubMed] [Google Scholar]
- 24.Campbell CL, Cadar D, McMunn A, Zaninotto P. Operationalization of Intrinsic Capacity in Older People and Its Association With Subsequent Disability, Hospital Admission and Mortality: Results From The English Longitudinal Study of Ageing. J Gerontol Biol Sci Med Sci. 2023;78(4):698–703. 10.1093/gerona/glac250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wakabayashi I, Daimon T. The cardiometabolic index as a new marker determined by adiposity and blood lipids for discrimination of diabetes mellitus. Clin Chim Acta. 2015;438:274–8. 10.1016/j.cca.2014.08.042. [DOI] [PubMed] [Google Scholar]
- 26.Zhou X, Tao X-L, Zhang L, Yang Q-K, Li Z-J, Dai L, et al. Association between cardiometabolic index and depression: National Health and Nutrition Examination Survey (NHANES) 2011–2014. J Affect Disord. 2024;351:939–47. 10.1016/j.jad.2024.02.024. [DOI] [PubMed] [Google Scholar]
- 27.Qiu W, Cai A, Li L, Feng Y. Association of depression trajectories and subsequent hypertension and cardiovascular disease: findings from the CHARLS cohort. J Hypertens. 2024;42:432–40. 10.1097/HJH.0000000000003609. [DOI] [PubMed] [Google Scholar]
- 28.Chen X, Crimmins E, Hu P (Perry), Kim JK, Meng Q, Strauss J Venous blood-based biomarkers in the China health and retirement longitudinal study, et al. editors. rationale, design, and results from the 2015 wave. Am J Epidemiol. 2019;188:1871–7. 10.1093/aje/kwz170. [DOI] [PMC free article] [PubMed]
- 29.Chew J, Lee J, Hernandez HHC, Munro YL, Lim CL, Lim WS. The vitality domain of intrinsic capacity: A scoping review of conceptual frameworks and measurements. J Frailty Aging. 2025;14(4):100058. 10.1016/j.tjfa.2025.100058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fuentealba M, Rouch L, Guyonnet S, Lemaitre JM, de Souto Barreto P, Vellas B, et al. A blood-based epigenetic clock for intrinsic capacity predicts mortality and is associated with clinical, immunological and lifestyle factors. Nat Aging. 2025;5(7):1207–16. 10.1038/s43587-025-00883-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.De Lima TR, González-Chica DA, Sui X, Santos Silva DA. The independent and joint associations among muscle strength, abdominal obesity and cardiometabolic variables among adults. Eur J Sport Sci. 2022;22(7):1122–31. 10.1080/17461391.2021.1902571. [DOI] [PubMed] [Google Scholar]
- 32.Patial R, Batta I, Thakur M, Sobti RC, Agrawal DK. Etiology, pathophysiology, and treatment strategies in the prevention and management of metabolic syndrome. Archives Intern Med Res. 2024;7(4):273–83. 10.29302/aimr.2024.070401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Byeon JY, Lee MK, Yu MS, Kang MJ, Lee DH, Kim KC, et al. Lower relative handgrip strength is significantly associated with a higher prevalence of the metabolic syndrome in adults. Metab Syndr Relat Disord. 2019;17(5):280–8. . 10.1089/met.2018.0111. [DOI] [PubMed] [Google Scholar]
- 34.D’Avila JC, Nabbout ME, Georges Moreira El Nabbout TG, Silva H, Barbosa Ramos Junior ADS, Fonseca AC. Correlation between low handgrip strength and metabolic syndrome in older adults: a systematic review. Arch Endocrinol Metab. 2024;68:e230026. 10.20945/2359-4292-2023-0026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wu H, Liu M, Chi VTQ, Wang J, Zhang Q, Liu L, et al. Handgrip strength is inversely associated with metabolic syndrome and its separate components in middle aged and older adults: A large-scale population-based Study . Metabolism. 2019;93:61–7. 10.1016/j.metabol.2019.01.011. [DOI] [PubMed] [Google Scholar]
- 36.Zhang Y, Du M, Li Z, Wang X, Leng M, Huang Y, et al. The Correlation Between Visceral Fat Area to Skeletal Muscle Mass Ratio and Multiorgan Insulin Resistance in Chinese Population With Obesity. Int J Endocrinol. 2024;2024:1297584. 10.1155/2024/. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Montoya Castillo M, Martínez Quiroz WJ, Suarez-Ortegón MF, Higuita-Gutiérrez LF. Waist-to-Height Ratio, Waist Circumference, and Body Mass Index in Relation to Full Cardiometabolic Risk in an Adult Population from Medellin, Colombia. J Clin Med. 2025;14(7):2411. 10.3390/jcm14072411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ko S-H, Jung Y. Energy Metabolism Changes and Dysregulated Lipid Metabolism in Postmenopausal Women. Nutrients. 2021;13(12):4556. 10.3390/nu13124556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Dam TV, Dalgaard LB, Thomsen CB, Hjortebjerg R, Ringgaard S, et al. Estrogen modulates metabolic risk profile after resistance training in early postmenopausal women: a randomized controlled trial. Menopause. 2021;28(11):1214–24. 10.1097/GME.0000000000001841. [DOI] [PubMed] [Google Scholar]
- 40.Kammerlander AA, Lyass A, Mahoney TF, Massaro JM, Long MT, et al. Sex Differences in the Associations of Visceral Adipose Tissue and Cardiometabolic and Cardiovascular Disease Risk: The Framingham Heart Study. J Am Heart Assoc. 2021;10(11):e019968. 10.1161/JAHA.120.019968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Meloni A, Cadeddu C, Cugusi L, Donataccio MP, Deidda M, et al. Gender Differences and Cardiometabolic Risk: The Importance of the Risk Factors. Int J Mol Sci. 2023;24(2):1588. 10.3390/ijms24021588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Lu X, Yao Y, Jin Y. Digital exclusion and functional dependence in older people: Findings from five longitudinal cohort studies. EClinicalMedicine. 2022;54:101708. 10.1016/j.eclinm.2022.101708. Published 2022 Oct 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Lv YB, Mao C, Gao X, Yin ZX, Kraus VB, Yuan JQ, et al. Triglycerides paradox among the oldest old: The lower the better? J Am Geriatr Soc. 2019;67(4):741–8. 10.1111/jgs.15733. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Kato K, Yokoyama H, Iwasaki T, Konno Y, Yamazaki K, Shikanai S, et al. Impact of Triglyceride Levels on the Long-term Clinical Outcomes in Patients With Acute Myocardial Infarction. Vivo. 2024;38(6):3078–84. 10.21873/invivo.13792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Sobów T, Fendler W, Magierski R. Body mass index and mild cognitive impairment-to-dementia progression in 24 months: a prospective study. Eur J Clin Nutr. 2014;68(11):1216–9. 10.1038/ejcn.2014.167. [DOI] [PubMed] [Google Scholar]
- 46.Chen J, Zhang J, Wang H, Zhu H, Fu J, Li C, et al. Obesity paradox in cognitive function: a longitudinal study of BMI and cognitive impairment in older adult Chinese population. Front Aging Neurosci. 2025;17(1543501). 10.3389/fnagi.2025.1543501. [DOI] [PMC free article] [PubMed]
- 47.Cruz-Pereira JS, Rea K, Nolan YM, O’Leary OF, Dinan TG, Cryan JF. Depression’s unholy trinity: dysregulated stress, immunity, and the microbiome. Annu Rev Psychol. 2020;71:49–78. 10.1146/annurev-psych-122216-011613. [DOI] [PubMed] [Google Scholar]
- 48.Ma N, Alifu J, Meng G, Ma J, Wang H, Meng X, Liu X. Uncovering the link between cardiometabolic index and depression in diabetes: a large-scale population study. Diabetol Metab Syndr. 2025;17(1):317. 10.1186/s13098-025-01881-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Peng KY, Chen ZJ, Tang FP, Lee SC, Tung HH, et al. Vitality attributes and their associations with intrinsic capacity, resilience, and happiness in community-dwelling adults: Results from Gan-Dau Healthy Longevity Plan Wave 2. J Nutr Health Aging. 2025;29(7):100559. 10.1016/j.jnha.2025.100559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.George PP, Lun P, Ong SP, Lim WS. A Rapid Review of the Measurement of Intrinsic Capacity in Older Adults. J Nutr Health Aging. 2021;25(6):774–82. 10.1007/s12603-021-1622-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Seidler RD, Bernard JA, Burutolu TB, Fling BW, Gordon MT, Gwin JT, et al. Motor control and aging: links to age-related brain structural, functional, and biochemical effects. Neurosci Biobehav Rev. 2010;34(5):721–33. 10.1016/j.neubiorev.2009.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Hoffstaedter F, Grefkes C, Roski C, Caspers S, Zilles K, Eickhoff SB. Age-related decrease of functional connectivity additional to gray matter atrophy in a network for movement initiation. Brain Struct Funct. 2015;220(2):999–1012. 10.1007/s00429-013-0696-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Wainman-Lefley J. Allostatic load and long-term outcome after head injury [dissertation]. Glasgow: University of Glasgow; 2017. Available from: https://heses.gla.ac.uk/7931/1/2017. Wainman-LefleyPhD.
- 54.Savitz DA, Wellenius GA. Can Cross-Sectional Studies Contribute to Causal Inference? It Depends. Am J Epidemiol. 2023;192(4):514–6. 10.1093/aje/kwac037. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the China Health and Retirement Longitudinal Study (CHARLS) repository. CHARLS data are publicly accessible upon reasonable request/registration through the official website: https://charls.pku.edu.cn/.