Abstract
Background
Cardiovascular disease (CVD) is the leading cause of morbidity and mortality worldwide. Intrinsic capacity (IC) represents a composite measure of physical and mental capacities. However, the relationships among baseline IC status, changes in IC over time, and the risk of incident CVD remain unclear.
Methods
This multicohort study used data from 4 nationally representative aging cohorts: the CHARLS (China Health and Retirement Longitudinal Study), the HRS (Health and Retirement Study), the MHAS (Mexican Health and Aging Study), and the SHARE (Survey of Health, Aging, and Retirement in Europe). Participants were aged ≥50 years. IC was assessed across 5 domains (locomotion, cognition, vitality, sensory, and psychology). Changes in IC status were categorized as robust, decline, improve, or stable impaired. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% CIs.
Results
A total of 4548 participants from CHARLS, 8420 from HRS, 8165 from MHAS, and 19 111 from SHARE were included in baseline IC status analyses. Compared with no IC decline, individuals with impairment in 2 or more domains had a significantly higher risk of incident CVD (CHARLS, HR=1.33, 95% CI: 1.12–1.58; HRS, HR=1.73, 95% CI: 1.48–2.02; MHAS, HR=1.31, 95% CI: 1.03–1.65; SHARE, HR=1.32, 95% CI: 1.19–1.46). A total of 2825 participants from CHARLS, 7981 from HRS, 6932 from MHAS, and 17 631 from SHARE were included in changes in IC status analyses. Participants with stable impaired (CHARLS, HR=1.65, 95% CI: 1.27–2.15; HRS, HR=1.57, 95% CI: 1.36–1.82; MHAS, HR=1.53, 95% CI: 1.09–2.14; SHARE, HR=1.53, 95% CI: 1.36–1.72) showed elevated CVD risks compared with the robust group.
Conclusions
Baseline IC impairments and sustaining impairments are associated with higher CVD risk. Routine assessment and monitoring of IC may prevent CVD.
Keywords: cardiovascular disease, intrinsic capacity, middle‐aged and older adults, multicohort
Subject Categories: Cardiovascular Disease, Aging
Nonstandard Abbreviations and Acronyms
- IC
intrinsic capacity
- CHARLS
China Health and Retirement Longitudinal Study
- HRS
Health and Retirement Study
- MHAS
Mexican Health and Aging Study
- SHARE
Survey of Health, Aging and Retirement in Europe
- TICS
Telephone Interview of Cognitive Status
Clinical Perspective.
What Is New?
Patients with baseline intrinsic capacity impairments and stable impaired are associated with increased cardiovascular disease risk.
What Are the Clinical Implications?
Incorporating intrinsic capacity evaluation into midlife and geriatric health screenings may help identify individuals at higher risk of cardiovascular disease and could be a valuable component of strategies aimed at promoting healthy aging and potentially mitigating cardiovascular disease risk.
Cardiovascular disease (CVD) is the leading cause of death and disability worldwide, responsible for >18 million deaths annually. 1 It disproportionately affects middle‐aged and older adults, and its prevalence rises steeply with advancing age. 2 Given the rapid pace of global population aging, the number of individuals living with or at risk of CVD is projected to surge to 45 million by 2050, 3 imposing a substantial financial burden on health systems. 4 , 5 , 6 This demographic shift underscores the urgent need for early detection strategies and preventive interventions. Identifying early physiological markers and modifiable risk factors is thus critical to mitigate future CVD burdens and guide public health strategies.
In response to global population aging, the World Health Organization introduced intrinsic capacity (IC) as the cornerstone of healthy aging framework. 7 IC encompasses an individual’s composite physical and mental reserves, shifting the focus from disease management to lifelong functional maintenance. IC comprises 5 key domains: locomotion, cognition, vitality, sensory function, and psychology. 8 , 9 Landmark longitudinal studies, such as ELSA (English Longitudinal Study of Aging) 10 and CHARLS (China Health and Retirement Longitudinal Study), 11 have validated IC as a robust predictor of future functional decline, independent of chronological age and comorbidities. Crucially, IC deterioration often precedes geriatric syndromes (eg, frailty) and disability onset. 12 Routine IC assessment in midlife therefore offers a pivotal window for early intervention to preserve function and improve long‐term outcomes.
Previous studies have demonstrated that IC is associated with the risk of incident CVD. 13 , 14 Cross‐sectional analyses further suggest that IC impairments are linked to alterations in inflammatory biomarkers, 15 , 16 which may promote endothelial dysfunction and accelerate atherosclerosis—both key precursors of CVD. However, these investigations focused only on baseline IC status, overlooking the dynamic changes in IC status over time. Additionally, the relationship between IC and CVD in the general population remains unclear. Compared with a single baseline measurement, evaluating longitudinal changes in IC may capture more comprehensive biological associations. Notably, emerging evidence indicates that IC can be reversed after appropriate intervention. 9 , 17 Understanding how changes in IC status are related to CVD risk could therefore provide evidence for developing targeted prevention strategies.
To address these knowledge gaps, we analyzed 4 prospective cohorts with nationally representative data from China, the United States, Mexico, and Europe. Our study aimed to examine the associations of baseline IC and changes in IC status with incident CVD events. We hypothesized that progression of IC impairments would increase incident CVD risk, while recovery of IC impairments status would reduce CVD risk.
METHODS
Data Availability
All original data used in the study from the CHARLS, HRS, MHAS, and SHARE can be freely downloaded from their official websites, which can be obtained from the Gateway to Global Aging Data (https://g2aging.org/home).
Study Design and Participants
In this secondary analysis of an observational cohort study, data were drawn from CHARLS, 18 HRS, 19 MHAS, 20 and SHARE. 21 These studies were designed with harmonized methodologies to facilitate cross‐national comparisons, providing information on intrinsic capacity (IC) and cardiovascular disease (CVD). To ensure consistent observation periods across cohorts, we included participants aged ≥50 years from waves 1–4 for CHARLS (2011–2018), waves 10–14 for HRS (2010–2018), waves 3–5 for MHAS (2012–2018), and waves 4–7 for SHARE (2010–2017).
In this study, wave 1 of CHARLS, wave 10 of HRS, wave 3 of MHAS, and wave 4 of SHARE were defined as baseline. Wave 2 of CHARLS, wave 11 of HRS, wave 4 of MHAS, and wave 5 of SHARE served as the second survey for assessing changes in IC status. Subsequent waves were used for outcome ascertainment until the final follow‐up.
The selection process for each cohort is detailed in the flowchart (Figure S1). Among 105 974 participants from CHARLS, HRS, MHAS, and SHARE, we excluded 13 188 participants with missing baseline IC and 52 542 participants with prevalent CVD at baseline or lost to follow‐up. This yielded 40 244 participants for baseline IC status analyses. For changes in IC analyses, we further excluded 4841participants according to similar criteria. The remaining 35 042 participants were included in the final analyses.
We used deidentified, publicly available data from 4 public databases. Ethical approval was approved by the original surveys, and no additional ethical approval was required for the present study. All participants provided written informed consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
Measurement of Intrinsic Capacity
IC was measured across 5 domains on the basis of the World Health Organization definition and prior research. 7 , 22 , 23 Each domain was scored from 0 to 1, with higher scores indicating better IC. The criteria for impairment in each domain were as follows: (1) locomotion: partial or complete need for assistance in getting up from the bed or from a chair after sitting for a long period; (2) vitality: body mass index (BMI) ≤18.5 kg/m2; (3) cognition: cognition assessed by episodic memory and mental status tests (eg, Telephone Interview of Cognitive Status or its modified versions, Telephone Interview of Cognitive Status ‐m and Telephone Interview of Cognitive Status‐10). To facilitate comparisons across cohorts, standardized z scores were constructed by subtracting the baseline mean and dividing by the SD. Cognitive impairment was defined as a cognition score at least 1 SD below the mean; (4) psychology: psychology assessed via the Center for Epidemiologic Studies Depression Scale and Euro‐Depression Scale: Center for Epidemiologic Studies Depression Scale‐8 in HRS and ELSA (range 0–8, threshold ≥3), Center for Epidemiologic Studies Depression Scale‐10 in CHARLS (range 0–30, threshold ≥10), Center for Epidemiologic Studies Depression Scale‐9 in MHAS (range 0–9, threshold ≥5), and Euro‐Depression Scale‐12 in SHARE (range 0–12, threshold ≥4). Scores above these thresholds indicated depressive symptoms 24 , 25 ; (5) sensory: self‐reported vision or hearing as “bad” or “very bad”. The Detailed measurement protocols for each cohort are elaborated in Table S1.
Individuals were identified as having impaired IC if they exhibited dysfunction in one or more of the 5 domains, and those with no impairments were categorized as no decline IC. 26 In the baseline analyses, IC was classified into 3 categories according to previous studies: no decline, one domain decline, and 2 or more domains decline. 27 Changes in IC status were evaluated by IC at baseline and during the second survey (Figure S2). We defined 4 changes in IC status: robust (no decline both in 2 surveys); decline (no decline at baseline while impaired in the second survey); improve (impaired at baseline while no decline in the second survey); and sustained impaired (impaired both in 2 surveys).
Outcome Assessment
The primary outcome of this study was incident CVD. In accordance with previous studies, 28 , 29 CVD was ascertained based on self‐reported physician‐diagnosed heart disease or stroke. Participants were asked in each wave “Has your doctor ever told you that you have a heart condition (including angina, heart attack, congestive heart failure and other heart problems)?” and “Have you doctor ever told you that you have been diagnosed with a stroke?”. Those who reported being diagnosed with heart disease or stroke were considered to have CVD. In the next wave, participants were required to confirm the existence of heart disease and stroke if they reported those in the last wave. If participants disputed self‐reported heart disease or stroke from previous waves, they were corrected retrospectively. Our CVD ascertainment was consistent with previous studies using the CHARLS, HRS, MHAS, and SHARE cohorts. 28 , 29 , 30 Details of the CVD ascertainment were described in Data S1.
In the baseline IC status analyses, follow‐up began at wave 1 of CHARLS, wave 10 of HRS, wave 3 of MHAS, and wave 4 of SHARE. In the changes in IC status analyses, follow‐up began at wave 2 of CHARLS, wave 11 of HRS, wave 4 of MHAS, and wave 5 of SHARE. The follow‐up time was calculated from the date of the baseline to the diagnosis of CVD, death or the end of follow‐up, whichever occurred first.
Covariates
Based on prior evidence and directed acyclic graph (DAG), the potential baseline covariates were considered included age, sex, BMI, systolic blood pressure (SBP, not available in SHARE), diastolic blood pressure (DBP, not available in SHARE), marital status, education, non‐pension net wealth, smoking, drinking, history of hypertension, diabetes, chronic lung disease, cancer, and depression (Figure S3). BMI was calculated by weight in kilograms divided by height in meters squared. For consistency among CHARLS, HRS, MHAS, and SHARE, marital status was divided into married or partnered and other marital status (separated, divorced, widowed, or never married). Education was classified as below high school, high school, and college or above. Labor force status was coded into 2 groups: unemployed, currently working or retired, based on questions regarding the participants’ current employment status. Non‐pension net wealth was stratified into tertiles (lowest to highest quintile). Smoking was categorized as current, former, and never smokers. In accordance with National Institute on Alcohol Abuse and Alcoholism guidelines, 31 participants were classified as engaging in excessive alcohol use if their calculated weekly consumption (drinking days/week multiplied by drinks/day) exceeded the sex‐specific thresholds of >14 drinks for men and > 7 for women. Participants that were not in this alcohol consumption range were classified as non‐excessive alcohol use. Details of other covariates were described in Data S1.
Statistical Analysis
Data were analyzed from December 2024 to June 2025. Mean and SD or median and interquartile range were calculated for continuous variables, and numbers and percentages were calculated for categorical variables. Normality was assessed via histograms. Missing covariate data are described in Tables S2‐S3. To account for missing data, we did multiple imputation by chained equations (10 imputations) considering all the covariates used in the main analysis. The estimates from each of the ten imputed datasets were combined into the pooled overall estimate with the use of Rubin’s rule. 32
To analyze the association of baseline IC status with the risks of incident CVD, Cox proportional hazards models were used to calculate the hazard ratio (HR) and 95% CI. Proportional hazards assumptions were confirmed by the Schoenfeld residual tests. Three models were fitted for the Cox regression, with no decline as the reference. Model 1 was unadjusted. Model 2 adjusted for age, sex, BMI, SPB and DBP. Model 3 further included marital status, education, non‐pension net wealth, labor force status, smoking, drinking, hypertension, diabetes, chronic lung disease, cancer, and depression as the minimal sufficient adjustment set (MSAS). Using similar methods, we also analyzed associations of changes in IC status with incident CVD, with robust as the reference. We further controlled for the baseline IC scores in model 3 in analyses of changes in IC status.
To ensure robustness, we performed several sensitivity analyses of IC at baseline and changes in IC status. First, we excluded participants with chronic comorbidities, including 2 or more of hypertension, diabetes, cancer, arthritis, lung disease, or memory disease at baseline because IC could be influenced by major chronic diseases. Second, we excluded who had incident CVD in the second years of follow‐up to avoid potential reverse causation bias. Third, IC was reconstructed considering different definitions in locomotion and vitality. Fourth, Cox analyses were further adjusted for medical insurance, activity of daily living, and physical activity. Fifth, we reran this analysis using the complete data set. Sixth, death was considered as a competitive risk of CVD and fitted a Fine–Gray subdistribution hazards regression model. Seventh, sex (men or women) and age‐stratified (<65 years or ≥ 65 years) analyses were conducted, and interaction terms were evaluated using likelihood ratio test. Eighth, an E‐value with 95% CI was calculated to evaluate the robustness of our primary outcome to unmeasured confounders based on the risk ratio scale. All analyses were conducted using R software (version 4.3.2). A 2‐sided P value <0.05 was considered statistically significant.
RESULTS
Baseline Characteristics
Baseline characteristics of included participants from the 4 cohorts were presented in Table 1. This cross‐national cohort study included 40 244 participants aged 50 years or older: 4548 from CHARLS (median age: 59.0 years; women: 45.7%), 8420 from HRS (median age: 61.0 years; women: 61.0%), 8165 from MHAS (median age: 63.0 years; women: 56.8%), and 19 111 from SHARE (median age: 63.0 years; women: 59.0%). Participants with ≥2 domain declines were more likely to be older, nondrinkers, have higher BMI, and higher prevalence of hypertension and diabetes (Tables S17 through S20). For changes in IC status analyses, 2858 participants from CHARLS (median age: 59.0 years; women: 42.5%), 7981 from HRS (median age: 61.0 years; women: 61.2%), 6932 from MHAS (median age: 62.0 years; women:55.1%), 17 631 from SHARE (median age: 63.0 years; women: 59.0%) were included according to corresponding criteria. Baseline characteristics of these participants were shown in Table 2. In addition, we describe the baseline characteristics between participants included and excluded (Tables S9 through S16).
Table 1.
Baseline Characteristics of Participants for Baseline IC Status in the CHARLS, HRS, MHAS, and SHARE Cohorts
| Characteristics | CHARLS | HRS | MHAS | SHARE |
|---|---|---|---|---|
| n=4548 | n=8420 | n=8165 | n=19 111 | |
| Age, y, median (IQR) | 59.0 (55.0–65.0) | 61.0 (55.0–70.0) | 63.0 (57.0–69.0) | 63.0 (57.0–70.0) |
| Female sex, n (%) | 2077 (45.7) | 5139 (61.0) | 4639 (56.8) | 11 272 (59.0) |
| BMI, mean±SD | 23.2±3.5 | 29.5±6.2 | 27.7±4.8 | 26.7±4.4 |
| SBP, mean±SD | 130.3±20.6 | 129.7±19.6 | 138.7±20.4 | / |
| DBP, mean±SD | 75.4±11.7 | 81.0±11.5 | 79.0±10.7 | / |
| Education, n (%) | ||||
| Below high school | 4053 (89.1) | 1308 (15.5) | 6988 (85.9) | 6830 (35.7) |
| High school | 460 (10.1) | 4411 (52.4) | 284 (3.5) | 7773 (40.7) |
| College or above | 35 (0.8) | 2699 (32.1) | 863 (10.6) | 4508 (23.6) |
| Marital status, n (%) | ||||
| Married and partnered | 4060 (89.3) | 5739 (68.2) | 5907 (72.3) | 13 432 (70.3) |
| Unmarried and others | 488 (10.7) | 2678 (31.8) | 2258 (27.7) | 5666 (29.7) |
| Labor force status, n (%) | ||||
| Unemployed | 123 (2.7) | 4405 (52.3) | 1690 (20.7) | 12 706 (67.4) |
| Employed | 4425 (97.3) | 4015 (47.7) | 6470 (79.3) | 6151 (32.6) |
| Net wealth, n (%) | ||||
| Low | 1022 (29.5) | 2589 (30.7) | 2539 (31.1) | 4787 (25.2) |
| Middle | 1245 (35.9) | 2779 (33.0%) | 2825 (34.6) | 6419 (33.8) |
| High | 1203 (34.7) | 3052 (36.2) | 2801 (34.3) | 7768 (40.9) |
| Smoking, n (%) | ||||
| Never | 2508 (55.1) | 3946 (47.1) | 5105 (62.5) | 10 472 (55.0) |
| Former | 413 (9.1) | 3239 (38.6) | 2042 (25.0) | 5040 (26.5) |
| Current | 1627 (35.8) | 1200 (14.3) | 1016 (12.4) | 3540 (18.6) |
| Drinking, n (%) | ||||
| None or light | 3606 (84.1) | 3138 (37.3) | 7921 (97.0) | 1642 (9.4) |
| Heavy | 681 (15.9) | 5282 (62.7) | 244 (3.0) | 15 738 (90.6) |
| Hypertension, n (%) | 1816 (40.2) | 4626 (54.9) | 4266 (52.2) | 7972 (41.7) |
| Diabetes, n (%) | 307 (6.8) | 1409 (16.7) | 1839 (22.5) | 2081 (10.9) |
| Lung disease, n (%) | 401 (8.8) | 403 (4.8) | 653 (8.0) | 1119 (5.9) |
| Cancer, n (%) | 27 (0.6) | 840 (10.0) | 211 (2.6) | 217 (2.9) |
| Depression, n (%) | 1440 (31.7) | 1513 (18.0) | 2403 (29.4) | 4538 (23.7) |
BMI indicates body mass index; CHARLS, China Health and Retirement Longitudinal Study; DBP, diastolic blood pressure; HRS, Health and Retirement Study; IQR, interquartile range; MHAS, Mexican Health and Aging Study; SBP, systolic blood pressure; and SHARE, Survey of Health, Aging and Retirement in Europe.
Table 2.
Baseline Characteristics of Participants for Changes in IC Status in the CHARLS, HRS, MHAS, and SHARE Cohorts
| CHARLS | HRS | MHAS | SHARE | |
|---|---|---|---|---|
| n=2858 | n=7981 | n=6932 | n=17 631 | |
| Age, y, median (IQR) | 59.0 (55.0–64.0) | 61.0 (55.0–70.0) | 62.0 (56.0–69.0) | 63.0 (57.0–70.0) |
| Female sex, n (%) | 1214 (42.5) | 4881 (61.2) | 3819 (55.1) | 10 407 (59.0) |
| BMI, mean±SD | 23.4±3.5 | 29.4±6.2 | 27.8±4.8 | 26.6±4.3 |
| SBP, mean±SD | 129.7±20.0 | 129.6±19.5 | 138.3±20.1 | / |
| DBP, mean±SD | 75.3±11.5 | 81.1±11.5 | 79.2±10.6 | / |
| Education, n (%) | ||||
| Below high school | 2504 (87.6) | 1198 (15.0) | 5830 (84.4) | 6121 (34.7) |
| High school | 327 (11.4) | 2544 (31.9) | 274 (4.0) | 7262 (41.2) |
| College or above | 27 (0.9) | 4237 (53.1%) | 801 (11.6) | 4248 (24.1) |
| Marital status, n (%) | ||||
| Married and partnered | 2587 (90.5) | 5454 (68.4) | 5100 (73.6) | 12 439 (70.6) |
| Unmarried and others | 271 (9.5) | 2525 (31.6) | 1832 (26.4) | 5179 (29.4) |
| Labor force status, n (%) | ||||
| Unemployed | 68 (2.4) | 4141 (51.9) | 1482 (21.4) | 11 513 (66.2) |
| Employed | 2790 (97.6) | 3840 (48.1) | 5447 (78.6) | 5888 (33.8) |
| Net wealth, n (%) | ||||
| Low | 601 (27.7) | 2413 (30.2) | 2059 (29.7) | 4296 (24.5) |
| Middle | 776 (35.8) | 2653 (33.2) | 2421 (34.9) | 5871 (33.5) |
| High | 792 (36.5) | 2915 (36.5) | 2452 (35.4) | 7339 (41.9) |
| Smoking, n (%) | ||||
| Never | 1534 (53.7) | 3744 (47.1) | 4223 (60.9) | 9608 (54.7) |
| Former | 270 (9.4) | 3079 (38.7) | 1816 (26.2) | 4679 (26.6) |
| Current | 1054 (36.9) | 1127 (14.2) | 891 (12.9) | 3292 (18.7) |
| Drinking, n (%) | ||||
| None or light | 1752 (61.3) | 2937 (36.8) | 5036 (72.7) | 1452 (9.0) |
| Heavy | 1106 (38.7) | 5044 (63.2) | 1894 (27.3) | 14 612 (91.0) |
| Hypertension, n (%) | 1102 (38.8) | 4335 (54.3) | 3548 (51.2) | 7157 (40.6) |
| Diabetes, n (%) | 202 (8.9) | 1306 (16.4) | 1536 (22.2) | 1825 (10.4) |
| Lung disease, n (%) | 245 (8.6) | 368 (4.6) | 542 (7.8) | 992 (5.6) |
| Cancer, n (%) | 14 (0.5) | 788 (9.9) | 163 (2.4) | 198 (2.9) |
| Depression, n (%) | 857 (30.0) | 1396 (17.5) | 1939 (28.0) | 4065 (23.1) |
BMI indicates body mass index; CHARLS, China Health and Retirement Longitudinal Study; DBP, diastolic blood pressure; HRS, Health and Retirement Study; IQR, interquartile range; MHAS, Mexican Health and Aging Study; SBP, systolic blood pressure; and SHARE, Survey of Health, Aging and Retirement in Europe.
Baseline IC Status and CVD
In the baseline IC status analyses, the median follow‐up periods were 7.0 years in the CHARLS, 7.8 years in the HRS, 6.1 years in the MHAS, and 6.0 years in the SHARE. During follow‐up, 939 participants from CHARLS, 1400 from HRS, 515 from MHAS, and 3039 from SHARE developed CVD (Tables S4 through S8). The Kaplan–Meier curves demonstrated that IC impaired participants had a higher risk of CVD (Figure S4). Table 3 displays the associations between baseline IC status and the risk of incident CVD across 4 cohorts. After adjusting for the MSAS, more than 2 domains decline participants had significantly elevated risks of incident CVD than no decline participants (CHARLS, HR=1.33, 95% CI: 1.12–1.58; HRS, HR=1.73, 95% CI: 1.48–2.02; MHAS, HR=1.31, 95% CI: 1.03–1.65; SHARE, HR=1.32, 95% CI: 1.19–1.46; Pooled, HR=1.52, 95% CI: 1.41–1.63). Country‐specific results also consistently indicated that more than 2 domains decline increased the risk of incident CVD, although no great difference was noted (Table S21).
Table 3.
Association of Baseline IC Status With Risks of Incident CVD
| Events/n | Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | ||
| CHARLS | |||||||
| No decline | 266/1477 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| 1 domain decline | 296/1496 | 1.11 (0.94–1.31) | 0.20 | 1.10 (0.93–1.30) | 0.21 | 1.12 (0.95–1.33) | 0.20 |
| 2+ domains decline | 377/1575 | 1.36 (1.16–1.59) | <0.001 | 1.32 (1.12–1.55) | <0.001 | 1.33 (1.12–1.58) | <0.001 |
| HRS | |||||||
| No decline | 640/4734 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| 1 domain decline | 493/2557 | 1.46 (1.30–1.64) | <0.001 | 1.38 (1.22–1.55) | <0.001 | 1.33 (1.18–1.50) | <0.001 |
| 2+ domains decline | 267/1129 | 1.89 (1.64–2.18) | <0.001 | 1.88 (1.63–2.19) | <0.001 | 1.73 (1.48–2.02) | <0.001 |
| MHAS | |||||||
| No decline | 226/4261 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| 1 domain decline | 163/2480 | 1.23 (1.01–1.51) | 0.04 | 1.18 (0.96–1.45) | 0.12 | 1.10 (0.89–1.35) | 0.39 |
| 2+ domains decline | 126/1424 | 1.64 (1.32–2.04) | <0.001 | 1.53 (1.22–1.93) | <0.001 | 1.31 (1.03–1.65) | 0.02 |
| SHARE | |||||||
| No decline | 1466/10862 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| 1 domain decline | 936/5571 | 1.24 (1.14–1.34) | <0.001 | 1.18 (1.09–1.28) | <0.001 | 1.14 (1.05–1.23) | 0.002 |
| 2+ domains decline | 637/2678 | 1.75 (1.60–1.93) | <0.001 | 1.43 (1.30–1.58) | <0.001 | 1.32 (1.19–1.46) | <0.001 |
| Pooled | |||||||
| No decline | 2598/21334 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| 1 domain decline | 1888/12104 | 1.28 (1.20–1.35) | <0.001 | 1.24 (1.16–1.31) | <0.001 | 1.20 (1.13–1.27) | <0.001 |
| 2+ domains decline | 1407/6806 | 1.71 (1.60–1.83) | <0.001 | 1.63 (1.53–1.74) | <0.001 | 1.52 (1.41–1.63) | <0.001 |
Model 1 no adjusted; Model 2 adjusted for age, sex, BMI, SBP, and DBP. Model 3 further adjusted education, marital status, labor force status, net wealth, smoking, drinking, hypertension, diabetes, lung disease, cancer, and depression. IC indicates intrinsic capacity; CVD, cardiovascular disease; CHARLS, China Health and Retirement Longitudinal Study; HR, hazard ratio; HRS, Health and Retirement Study; MHAS, Mexican Health and Aging Study; and SHAR, Survey of Health, Aging and Retirement in Europe.
Changes in IC Status and CVD
Among baseline no decline IC participants, 526 (18.4%) from CHARLS, 964 (12.1%) from HRS, 1159 (16.7%) from MHAS, and 2387 (13.5%) from SHARE progressed to IC impaired. Meanwhile, among baseline IC impaired participants, 478 (16.7%), 977 (12.2%), 997 (14.4%), and 3193 (18.1%) recovered to no decline in CHARLS, HRS, MHAS, and SHARE, respectively (Table S22). Baseline characteristics by changes in IC status were reported in Tables S23‐S26.
In the changes in IC status analyses, the median follow‐up periods were 5.0 years in the CHARLS, 6.0 years in the HRS, 3.1 years in the MHAS, and 6.0 years in the SHARE, respectively. The Kaplan–Meier curves indicated that stable impaired participants had a higher risk of CVD (Figure S5). Table 4 shows the association between changes in IC status and risks of incident CVD. Compared with robust participants, stable impaired showed significantly elevated risks of incident CVD (CHARLS, HR=1.65, 95% CI: 1.27–2.15; HRS, HR=1.57, 95% CI: 1.36–1.82; MHAS, HR=1.53, 95% CI: 1.09–2.14; SHARE, HR=1.53, 95% CI: 1.36–1.72; Pooled, HR=1.64, 95% CI: 1.51–1.78). Improved participants also presented d elevated risks of incident CVD (CHARLS, HR=1.46, 95% CI: 1.08–1.98; HRS, HR=1.31, 95% CI: 1.07–1.58; Pooled, HR=1.18, 95% CI: 1.07–1.31), but not in MHAS (HR=0.87, 95% CI: 0.54–1.41). Country‐specific findings were broadly consistent, with persistent impairment showing the most pronounced risk (Table S27).
Table 4.
Association of Changes in IC Status With Risks of Incident CVD
| Events/n | Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | ||
| CHARLS | |||||||
| Robust | 84/597 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| Decline | 76/447 | 1.22 (0.89–1.66) | 0.21 | 1.26 (0.92–1.71) | 0.13 | 1.33 (0.97–1.82) | 0.08 |
| Improve | 90/478 | 1.35 (1.00–1.82) | 0.05 | 1.35 (1.00–1.84) | 0.05 | 1.46 (1.08–1.98) | 0.01 |
| Stable impaired | 276/1336 | 1.51 (1.18–1.93) | 0.001 | 1.57 (1.22–2.02) | <0.001 | 1.65 (1.27–2.15) | <0.001 |
| HRS | |||||||
| Robust | 381/3580 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| Decline | 122/964 | 1.22 (0.99–1.49) | 0.06 | 1.16 (0.94–1.42) | 0.15 | 1.14 (0.93–1.41) | 0.21 |
| Improve | 143/977 | 1.37 (1.13–1.66) | 0.001 | 1.33 (1.09–1.62) | 0.005 | 1.31 (1.07–1.58) | 0.007 |
| Stable impaired | 428/2460 | 1.71 (1.49–1.97) | <0.001 | 1.63 (1.41–1.88) | <0.001 | 1.57 (1.36–1.82) | <0.001 |
| MHAS | |||||||
| Robust | 66/2590 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| Decline | 45/1159 | 1.55 (1.06–2.26) | 0.02 | 1.48 (1.01–2.16) | 0.04 | 1.35 (0.92–1.99) | 0.13 |
| Improve | 24/997 | 0.95 (0.60–1.53) | 0.84 | 0.95 (0.59–1.53) | 0.84 | 0.87 (0.54–1.41) | 0.57 |
| Stable impaired | 107/2186 | 1.90 (1.40–2.59) | <0.001 | 1.81 (1.30–2.52) | <0.001 | 1.53 (1.09–2.14) | 0.01 |
| SHARE | |||||||
| Robust | 639/7837 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| Decline | 295/2387 | 1.60 (1.40–1.84) | <0.001 | 1.55 (1.35–1.78) | <0.001 | 1.50 (1.31–1.73) | <0.001 |
| Improve | 304/3193 | 1.13 (0.98–1.29) | 0.15 | 1.10 (0.96–1.26) | 0.18 | 1.06 (0.93–1.22) | 0.38 |
| Stable impaired | 611/4214 | 1.81 (1.62–2.02) | <0.001 | 1.65 (1.46–1.85) | <0.001 | 1.53 (1.36–1. 72) | <0.001 |
| Pooled | |||||||
| Robust | 1170/14604 | Reference | 1.00 | Reference | 1.00 | Reference | 1.00 |
| Decline | 538/4957 | 1.53 (1.38–1.69) | <0.001 | 1.50 (1.35–1.66) | <0.001 | 1.41 (1.28–1.57) | <0.001 |
| Improve | 561/5645 | 1.25 (1.13–1.38) | <0.001 | 1.23 (1.12–1.37) | <0.001 | 1.18 (1.07–1.31) | 0.004 |
| Stable impaired | 1422/10196 | 1.95 (1.81–2.11) | <0.001 | 1.89 (1.74–2.05) | <0.001 | 1.64 (1.51–1.78) | <0.001 |
Model 1 no adjusted; Model 2 adjusted for baseline IC, age, sex and BMI. Model 3 further adjusted for education, marital status, net wealth, smoking, drinking, hypertension, diabetes, lung disease, cancer, and depression. IC indicates intrinsic capacity; CVD, cardiovascular disease; CHARLS, China Health and Retirement Longitudinal Study; HR, hazard ratio; HRS, Health and Retirement Study; MHAS, Mexican Health and Aging Study; and SHARE, Survey of Health, Aging and Retirement in Europe.
Sensitivity Analyses
The association between baseline IC status and changes in IC status with incident CVD events remained consistent in the following analyses: when excluding participants with chronic comorbidities (Tables S28 through S29); when excluding participants with incident CVD during the second year of follow‐up (Tables S30 through S31); when reconstructing IC (Tables S32 through S33); after further adjusting for medical insurance, activity of daily living, and physical activity (Tables S34 through S35); when using the complete dataset (Tables S36 through S37); when considering death as a competitive risk of CVD (Tables S38 through S39). Results were also consistent in analyses stratified by sex and age groups (Tables S40 through S43). Of note, stronger effect size was observed in women compared with men, although sex interaction was not statistically significant. E‐value calculations suggested that only a confounder with a strong association with both exposure and outcome could fully explain away the observed associations (Tables S44 through S45).
DISCUSSION
To our knowledge, this is the first multinational cohort study to evaluate the associations of baseline and changes in IC status with incident CVD. We found that IC impairment participants were associated with elevated risks of CVD than no IC decline. Moreover, both stable impaired and decline showed elevated risks of CVD than robust participants. These findings were consistent across major aging cohorts, underscoring the importance of monitoring or restoring IC in long‐term CVD prevention.
As global aging trends continue, the prevalence of CVD among older adults is rising, imposing a substantial burden on healthcare systems and representing a growing public health concern. The strong link between IC status and cardiovascular health has been documented. 33 , 34 , 35 A large‐scale cohort using UK Biobank data found that IC deficits increased incidence risks of CVD and subsequent CVD mortality. 13 Song et al 36 found that a positive association between baseline IC and its longitudinal changes and incident CVD risk among individuals with cardiovascular‐kidney‐metabolic syndrome. Our study demonstrated that baseline IC impairments was associated with higher risks of incident CVD compared with no decline in participants. Recent research suggests that biological mechanisms linking IC with CVD might be shared pathophysiological pathways, including systemic inflammation, endothelial dysfunction, and neurohormonal activation. 37 , 38 Nevertheless, our results remained statistically significant after adjusting for traditional risk factors of CVD, suggesting that IC may serve as a potentially independent marker of CVD risk. These findings support the hypothesis that IC could be a useful target for risk stratification, highlighting the potential value of incorporating IC assessment into broader strategies for cardiovascular prevention. However, clinical utility would require validation in future interventional studies.
In addition to baseline IC status, we further examined the associations between changes in IC status and CVD risk, which were not investigated previously. 39 The primary limitations of prior epidemiologic studies on IC and CVD outcomes are single timepoint measures. Evidence indicates that IC is a dynamic process that can improve, deteriorate, or remain stable over time at critical periods of aging. 40 , 41 , 42 In addition, although no causal conclusion can be drawn, changes in IC status related to CVD reinforce the robustness and the biological plausibility of the reported associations. The highest risk of incident CVD was observed in participants with stable impairment, with HR of 1.89 (95% CI, 1.74–2.05) in the pooled data. Similar findings from longitudinal studies in diverse populations have demonstrated accumulation of IC deficits significantly elevated the risk of cardiovascular events and adverse outcomes. 26 , 42 , 43 Furthermore, participants with improved IC from impaired to robust also had an elevated risk of incident CVD compared with robust counterparts. This finding highlighted the prominence of achieving as early as possible in life a high level of IC and the importance of maintaining a robust IC over time. Evidence suggests that exposure to risk factors early in life is associated with premature CVD and mortality in adulthood and that early years of life play a significant part in influencing behaviors in adulthood. 44 We strongly suggest integrating the World Health Organization Integrated Care for Older People framework into CVD management protocols.
IC comprised 5 core domains: movement, vitality, hearing, vision, cognition, and psychology. Unlike conventional organ‐specific health assessments, IC reflects the integrated function of multiple physiological systems, providing a more holistic measure of an individual’s physiological reserve. This comprehensive approach is particularly relevant in the context of CVD, where the complex interplay between modifiable risk factors and vascular pathophysiology. Our findings have important clinical and public health implications. First, routine assessment of IC could enhance risk stratification in CVD, particularly for identifying individuals who might benefit from more intensive monitoring or intervention. No decline individuals also need to evaluate IC so that at‐risk individuals can be identified early and tailored prevention measures can be performed. Second, multimodal interventions (eg, structured exercise training, cognitive stimulation, nutritional optimization, and management of multimorbidity) might modify cardiovascular risk. These findings highlight the potential value of IC‐based approaches in CVD populations, though future randomized controlled trials are needed to establish causal efficacy and optimize intervention strategies.
Despite the robust and consistent associations observed, several limitations warrant consideration. First, CVD ascertainment was based on self‐reported physician diagnoses, which might lead to a misclassification bias, although we mitigated this by confirming diagnoses in subsequent waves. However, a previous study found that that the misreporting of stroke was non‐systematic, and self‐reported stroke could be used to study stroke incidence and risk factors in the HRS. 45 Second, the harmonization of measures across international cohorts, while enabling cross‐national comparisons, may have masked subtle cultural or methodological differences in IC assessment. Third, residual confounding cannot be entirely excluded, despite the E‐value being calculated. Fourth, selection bias might occur for changes in IC status analyses as we further excluded 5202 participants from the baseline IC status analyses. We compared the baseline characteristics of included and excluded participants (Tables S13‐S16), indicating the existence of healthy participant bias. Fifth, as an observational study, the reported associations between IC and incident CVD cannot be interpreted as being causal. Furthermore, the IC score assigned the same weight to each metric, whereas some may have stronger associations with CVD. Two points in time may lack precision to reliably estimate change in IC status over time. Finally, while the study includes multiple cohorts, cross‐national comparisons are still challenging because of differences in healthcare infrastructure, access to care, and health behaviors.
CONCLUSIONS
Baseline IC impairments and stable impairment status are associated with higher risk of incident CVD. Early identification and proactive management of IC impairments may be critical for CVD prevention.
Sources of Funding
This study was supported by National Natural Science Foundation of China (No. 72074168; No. 72304210) and Project for the Promotion and Optimization of Diagnostic and Therapeutic Technologies in Municipal Hospitals, Shanghai Shenkang Hospital Development Center (SHDC22024207).
Disclosures
None.
Supporting information
Supplemental Methods S1
Tables S1–S45
Figures S1–S5
Acknowledgments
We thank the Gateway to Global Aging Data for providing harmonized data information. We also appreciate the work done by staff and investigators from the China Health and Retirement Longitudinal Study; the Health and Retirement Study; the Mexican Health and Aging Study; and the Survey of Health, Aging and Retirement in Europe, and we appreciate the participation of respondents in these studies. Author contributions: LT and CQ designed the research, applied for the original data, and supervised the study. LT, CQ, and LY performed data collection, statistical analysis, and drafted the original manuscript. Fl and LW contributed to data visualization and manuscript revision.
This manuscript was sent to Jose R. Romero, MD, Associate Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.125.045986
For Sources of Funding and Disclosures, see page 9.
References
- 1. Virani SS, Alonso A, Benjamin EJ, Bittencourt MS, Callaway CW, Carson AP, Chamberlain AM, Chang AR, Cheng S, Delling FN, et al. Heart disease and stroke Statistics‐2020 update: a report from the American Heart Association. Circulation. 2020;141:e139–e596. doi: 10.1161/cir.0000000000000757 [DOI] [PubMed] [Google Scholar]
- 2. Global, regional, and National Burden of cardiovascular diseases and risk factors in 204 countries and territories, 1990‐2023. J Am Coll Cardiol. 2025;86:2167–2243. doi: 10.1016/j.jacc.2025.08.015 [DOI] [PubMed] [Google Scholar]
- 3. Joynt Maddox KE, Elkind MSV, Aparicio HJ, Commodore‐Mensah Y, de Ferranti SD, Dowd WN, Hernandez AF, Khavjou O, Michos ED, Palaniappan L, et al. Forecasting the burden of cardiovascular disease and stroke in the United States through 2050‐prevalence of risk factors and disease: a presidential advisory from the American Heart Association. Circulation. 2024;150:e65–e88. doi: 10.1161/cir.0000000000001256 [DOI] [PubMed] [Google Scholar]
- 4. Khan MS, Shahid I, Bennis A, Rakisheva A, Metra M, Butler J. Global epidemiology of heart failure. Nat Rev Cardiol. 2024;21:717–734. doi: 10.1038/s41569-024-01046-6 [DOI] [PubMed] [Google Scholar]
- 5. Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS, Boehme AK, Buxton AE, Carson AP, Commodore‐Mensah Y, et al. Heart disease and stroke Statistics‐2022 update: a report from the American Heart Association. Circulation. 2022;145:e153–e639. doi: 10.1161/cir.0000000000001052 [DOI] [PubMed] [Google Scholar]
- 6. Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, Barengo NC, Beaton AZ, Benjamin EJ, Benziger CP, et al. Global burden of cardiovascular diseases and risk factors, 1990–2019: update from the GBD 2019 study. J Am Coll Cardiol. 2020;76:2982–3021. doi: 10.1016/j.jacc.2020.11.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Beard JR, Officer A, de Carvalho IA, Sadana R, Pot AM, Michel JP, Lloyd‐Sherlock P, Epping‐Jordan JE, Peeters G, Mahanani WR, et al. The world report on ageing and health: a policy framework for healthy ageing. Lancet. 2016;387:2145–2154. doi: 10.1016/s0140-6736(15)00516-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Cesari M, Araujo de Carvalho I, Amuthavalli Thiyagarajan J, Cooper C, Martin FC, Reginster JY, Vellas B, Beard JR. Evidence for the domains supporting the construct of intrinsic capacity. J Gerontol A Biol Sci Med Sci. 2018;73:1653–1660. doi: 10.1093/gerona/gly011 [DOI] [PubMed] [Google Scholar]
- 9. WHO Guidelines Approved by the Guidelines Review Committee . Integrated Care for Older People: Guidelines on Community‐Level Interventions to Manage Declines in Intrinsic Capacity. Geneva: World Health Organization Copyright © World Health Organization 2017; 2017. [PubMed] [Google Scholar]
- 10. Beard JR, Jotheeswaran AT, Cesari M, Araujo de Carvalho I. The structure and predictive value of intrinsic capacity in a longitudinal study of ageing. BMJ Open. 2019;9:e026119. doi: 10.1136/bmjopen-2018-026119 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. 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 A Biol Sci Med Sci. 2022;77:94–100. doi: 10.1093/gerona/glab226 [DOI] [PubMed] [Google Scholar]
- 12. Liu S, Kang L, Liu X, Zhao S, Wang X, Li J, Jiang S. Trajectory and correlation of intrinsic capacity and frailty in a Beijing elderly community. Front Med (Lausanne). 2021;8:751586. doi: 10.3389/fmed.2021.751586 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Ramírez‐Vélez R, Iriarte‐Fernández M, Santafé G, Malanda A, Beard JR, Garcia‐Hermoso A, Izquierdo M. Association of intrinsic capacity with incidence and mortality of cardiovascular disease: prospective study in UK biobank. J Cachexia Sarcopenia Muscle. 2023;14:2054–2063. doi: 10.1002/jcsm.13283 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Guo Z, Chen Y, Koirala B, Li J, Jia Y, Dennison Himmelfarb CR, Jin J. Intrinsic capacity trajectories and cardiovascular disease incidence among Chinese older adults: a population‐based prospective cohort study. BMC Geriatr. 2025;25:269. doi: 10.1186/s12877-025-05910-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Chen ZJ, Lu WH, Meng LC, Chao WF, Tung HH, Hsiao FY, Chen LK. Intrinsic capacity impairments (ICOPE step 1 and step 2), Cardiometabolic risk and immune resilience: an exploratory analysis from the Gan‐Dau healthy longevity plan. J Frailty Aging. 2024;13:541–551. doi: 10.14283/jfa.2024.66 [DOI] [PubMed] [Google Scholar]
- 16. Lu WH, Gonzalez‐Bautista E, Guyonnet S, Lucas A, Parini A, Walston JD, Vellas B, de Souto Barreto P. Plasma inflammation‐related biomarkers are associated with intrinsic capacity in community‐dwelling older adults. J Cachexia Sarcopenia Muscle. 2023;14:930–939. doi: 10.1002/jcsm.13163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wang Y, Tang N, Shao M, Song J, Su Q, Gao Y. Transitions and trajectories in intrinsic capacity states over time: a systematic review. BMC Geriatr. 2025;25:407. doi: 10.1186/s12877-025-05938-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China health and retirement longitudinal study (CHARLS). Int J Epidemiol. 2012;43:61–68. doi: 10.1093/ije/dys203 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JW, Weir DR. Cohort profile: the health and retirement study (HRS). Int J Epidemiol. 2014;43:576–585. doi: 10.1093/ije/dyu067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Wong R, Michaels‐Obregon A, Palloni A. Cohort profile: the Mexican health and aging study (MHAS). Int J Epidemiol. 2015;46:e2. doi: 10.1093/ije/dyu263 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Börsch‐Supan A, Brandt M, Hunkler C, Kneip T, Korbmacher J, Malter F, Schaan B, Stuck S, Zuber S; Team obotSCC . Data resource profile: the survey of health, ageing and retirement in Europe (SHARE). Int J Epidemiol. 2013;42:992–1001. doi: 10.1093/ije/dyt088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Yu R, Lai ETC, Leung G, Ho SC, Woo J. Intrinsic capacity and 10‐year mortality: findings from a cohort of older people. Exp Gerontol. 2022;167:111926. doi: 10.1016/j.exger.2022.111926 [DOI] [PubMed] [Google Scholar]
- 23. Chen M, Hanewald K, Si Y, Gu Y, Beard JR. Intrinsic capacity across 15 countries in the survey of health, aging, and retirement in Europe. JAMA Netw Open. 2025;8:e259792‐e259792. doi: 10.1001/jamanetworkopen.2025.9792 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Radloff LS. The CES‐D scale:a self‐report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. doi: 10.1177/014662167700100306 [DOI] [Google Scholar]
- 25. Maskileyson D, Seddig D, Davidov E. The EURO‐D measure of depressive symptoms in the aging population: comparability across European countries and Israel. Front Polit Sci. 2021;3. doi: 10.3389/fpos.2021.665004 [DOI] [Google Scholar]
- 26. Zhang M, Xu Y, Xing Y, Li H. Association between multimorbidity and intrinsic capacity among older Chinese adults: evidence from the CHARLS 2011‐2015. Eur Geriatr Med. 2025;16:1207–1216. doi: 10.1007/s41999-025-01232-w [DOI] [PubMed] [Google Scholar]
- 27. Pan Y, Li X, Zhang L, Li Y, Tang Z, Ma L. Declined intrinsic capacity predicts long‐term mortality in Chinese older adults: Beijing longitudinal study of aging. Maturitas. 2024;188:108082. doi: 10.1016/j.maturitas.2024.108082 [DOI] [PubMed] [Google Scholar]
- 28. He D, Wang Z, Li J, Yu K, He Y, He X, Liu Y, Li Y, Fu R, Zhou D, et al. Changes in frailty and incident cardiovascular disease in three prospective cohorts. Eur Heart J. 2024;45:1058–1068. doi: 10.1093/eurheartj/ehad885 [DOI] [PubMed] [Google Scholar]
- 29. Zhang Z, Xu H, Zhang R, Yan Y, Ling X, Meng Y, Zhang X, Wang Y. Frailty and depressive symptoms in relation to cardiovascular disease risk in middle‐aged and older adults. Nat Commun. 2025;16:6008. doi: 10.1038/s41467-025-61089-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Zeng Q, Zhao L, Zhong Q, An Z, Li S. Changes in sarcopenia and incident cardiovascular disease in prospective cohorts. BMC Med. 2024;22:607. doi: 10.1186/s12916-024-03841-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Gunzerath L, Faden V, Zakhari S, Warren K. National Institute on Alcohol Abuse and Alcoholism report on moderate drinking. Alcohol Clin Exp Res. 2004;28:829–847. doi: 10.1097/01.alc.0000128382.79375.b6 [DOI] [PubMed] [Google Scholar]
- 32. Raghunathan TE. What do we do with missing data? Some options for analysis of incomplete data. Annu Rev Public Health. 2004;25:99–117. doi: 10.1146/annurev.publhealth.25.102802.124410 [DOI] [PubMed] [Google Scholar]
- 33. O’Donnell M, Teo K, Gao P, Anderson C, Sleight P, Dans A, Marzona I, Bosch J, Probstfield J, Yusuf S. Cognitive impairment and risk of cardiovascular events and mortality. Eur Heart J. 2012;33:1777–1786. doi: 10.1093/eurheartj/ehs053 [DOI] [PubMed] [Google Scholar]
- 34. Gao JW, Hao QY, Lu LY, Han JJ, Huang FF, Vuitton DA, Wang JF, Zhang SL, Liu PM. Associations of long‐term physical activity trajectories with coronary artery calcium progression and cardiovascular disease events: results from the CARDIA study. Br J Sports Med. 2022;56:854–861. doi: 10.1136/bjsports-2021-105092 [DOI] [PubMed] [Google Scholar]
- 35. Vaccarino V, Bremner JD. Stress and cardiovascular disease: an update. Nat Rev Cardiol. 2024;21:603–616. doi: 10.1038/s41569-024-01024-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Song F, Hu MJ, Xia J, Yin C. Intrinsic capacity and its change predict cardiovascular disease risk in early cardiovascular‐kidney‐metabolic syndrome: a nationwide cohort study. Open Heart. 2025;12. doi: 10.1136/openhrt-2025-003395 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Lu WH, González‐Bautista E, Guyonnet S, Martinez LO, Lucas A, Parini A, Rolland Y, Vellas B, de Souto Barreto P. Investigating three ways of measuring the intrinsic capacity domain of vitality: nutritional status, handgrip strength and ageing biomarkers. Age Ageing. 2023;52. doi: 10.1093/ageing/afad133 [DOI] [PubMed] [Google Scholar]
- 38. Meng LC, Huang ST, Peng LN, Chen LK, Hsiao FY. Biological features of the outcome‐based intrinsic capacity composite scores from a population‐based cohort study: pas de Deux of biological and functional aging. Front Med (Lausanne). 2022;9:851882. doi: 10.3389/fmed.2022.851882 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Sánchez‐Sánchez JL, Lu WH, Gallardo‐Gómez D, Del Pozo Cruz B, de Souto Barreto P, Lucia A, Valenzuela PL. Association of intrinsic capacity with functional decline and mortality in older adults: a systematic review and meta‐analysis of longitudinal studies. Lancet Healthy Longev. 2024;5:e480–e492. doi: 10.1016/s2666-7568(24)00092-8 [DOI] [PubMed] [Google Scholar]
- 40. Jia S, Zhao W, Ge M, Xia X, Hu F, Hao Q, Zhang Y, Yang M, Yue J, Dong B. Associations between transitions of intrinsic capacity and frailty status, and 3‐year disability. BMC Geriatr. 2023;23:96. doi: 10.1186/s12877-023-03795-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Cho HJ, Jung H, Won CW, Kim M. Intrinsic capacity transitions as predictors of frailty transitions in community‐dwelling older adults: findings from the Korean frailty and aging cohort study. Arch Gerontol Geriatr. 2025;128:105637. doi: 10.1016/j.archger.2024.105637 [DOI] [PubMed] [Google Scholar]
- 42. Hwang AC, Chen LY, Tseng SH, Huang CY, Yen KH, Chen LK, Lin MH, Peng LN. Intrinsic capacity transitions predict overall and cause‐specific mortality, incident disability, and healthcare utilization. J Nutr Health Aging. 2024;28:100359. doi: 10.1016/j.jnha.2024.100359 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Zhao M, Wu C, Sha S, Si Y, Peng S, Li M, Wang K. Longitudinal transitions in intrinsic capacity profiles and their associations with subsequent adverse outcomes among Chinese older adults: a nationwide prospective cohort study. Geriatr Nurs. 2025;64:103351. doi: 10.1016/j.gerinurse.2025.04.024 [DOI] [PubMed] [Google Scholar]
- 44. Tan LF, How AL, Lee XX, Tan BYQ, Lim YW, Lee L, Khoo SS, Jia L, Merchant RA. Intrinsic capacity deficits across the lifespan in a nationally representative community cohort: findings from the Queenstown study. Arch Gerontol Geriatr. 2025;140:106031. doi: 10.1016/j.archger.2025.106031 [DOI] [PubMed] [Google Scholar]
- 45. Xie W, Zheng F, Yan L, Zhong B. Cognitive decline before and after incident coronary events. J Am Coll Cardiol. 2019;73:3041–3050. doi: 10.1016/j.jacc.2019.04.019 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Methods S1
Tables S1–S45
Figures S1–S5
Data Availability Statement
All original data used in the study from the CHARLS, HRS, MHAS, and SHARE can be freely downloaded from their official websites, which can be obtained from the Gateway to Global Aging Data (https://g2aging.org/home).