Abstract
Objective
Association between intrinsic capacity (IC) and venous thromboembolism (VTE) remains unclear, we aim to explore the relationship among IC, genetic predisposition, and incident VTE.
Methods
In this large retrospective cohort study using the UK Biobank, IC, defined as the composite of all physical and mental capacities of an individual, was assessed across 7 factors (such as exhaustion, sleep duration, eye problems, hearing difficulty, weight loss, gait speed, and grip strength) and scored from 0 (best) to 7 (worst). Participants were grouped into five IC categories: 0, 1, 2, 3, and ≥4 factors. Incident VTE was identified through national medical records. Genetic predisposition to VTE was determined by the tertiles of polygenic risk score (PRS) derived from external genome-wide association data from the UK Biobank. Cox proportional hazards models were used to investigate the associations with multiple variables adjusted.
Results
A total of 415,581 participants (median age, 57.0 years; interquartile range, 50.0-63.0; 54.96% female; 94.89% White) were included in the final analysis. Over a median follow-up of 14.37 years, 8299 (2.00%) developed VTE, including 3878 (0.94%) with and 4421 (1.07%) without a history of cancer. Higher IC scores were associated with increased VTE risk compared to those with IC = 0 (for IC = 1: hazard ratio (HR), 1.08; 95% confidence interval (CI), 1.01-1.16; P = .029; for IC = 2: HR, 1.12; 95% CI, 1.04-1.22; P = .005; for IC = 3: HR, 1.19; 95% CI, 1.06-1.34; P = .003; for IC ≥ 4: HR, 1.46; 95% CI, 1.23-1.75; P < .001). Similar associations were observed in both cancer and noncancer subgroups. Furthermore, individuals with both the highest PRS and the worst IC (≥4 factors) had the greatest VTE risk, with significant interactions between PRS and IC.
Conclusions
Our large-scale study suggests that poor IC is an independent risk factor for VTE. Moreover, integrating IC with genetic risk can identify a high-risk subgroup, which supports the clinical utility of IC assessment in VTE risk stratification.
Keywords: Venous thromboembolism, Intrinsic capacity, Poly genetic risk score, Risk stratification
Article Highlights.
-
•
Type of Research: Retrospective population-based cohort study
-
•
Key Findings: Intrinsic capacity (IC) is the composite of all physical and mental capacities of an individual. Among 415,581 participants, worse IC was associated with higher risk of venous thromboembolism (VTE) (hazard ratio [95% confidence interval]: 1.08 [1.01-1.16], 1.12 [1.04-1.22], 1.19 [1.06-1.34], and 1.46 [1.23-1.75] for IC scores of 1 to ≥4). Polygenic risk scores are now widely applied to evaluate individual genetic susceptibility to disease risk. Participants with the highest polygenic risk scores of VTE and the worst IC had the greatest VTE risk.
-
•
Take Home Message: Worse IC status was related to elevated VTE risk. IC may serve as a feasible tool for VTE risk stratification given its noninvasive and rapid measures.
Venous thromboembolism (VTE), including deep vein thrombosis and pulmonary embolism, is the third leading cause of cardiovascular mortality.1 With the global demographic transition toward population ageing, the burden of VTE is projected to rise sharply, as older age is one of the strongest independent risk factors for VTE.2 This trend highlights the urgency of developing cost-effective biomarkers to enable early identification, stratification, and prevention of VTE.
In response to the challenges posed by ageing population, the World Health Organization (WHO) introduced intrinsic capacity (IC) in its 2015 World Report on Ageing and Health as a central pillar in advancing healthy ageing.3 IC, which is defined as “the composite of all the physical and mental capacities that an individual can draw on at any point in time throughout life,” is a dynamic indicator, emphasizing function over the mere presence or the absence of disease.4,5 Recent longitudinal studies have demonstrated that declines in IC are associated with higher risk of mortality, disability and age-related diseases such as cardiovascular disease, dementia, sarcopenia, and mortality.6, 7, 8, 9, 10, 11, 12 These findings suggest that IC may serve as a dynamic biomarker of biological ageing and functional resilience. However, despite well-established links between IC and cardiovascular disease, its potential association with VTE remains unexplored. Elucidating if IC is related to VTE risk would aid in early identification of VTE and help explore the role of IC in healthy ageing managing. Given that VTE and IC are both closely tied to ageing and chronic physiological decline,6,13,14 investigating the relationship between IC status and VTE risk could provide valuable insights for early risk stratification and VTE prevention within ageing populations.
The ageing context also intersects with cancer, another major health challenge. In patients with cancer, VTE is a major contributor to mortality and worse clinical outcomes.15,16 Given the high prevalence and poor prognosis associated with VTE, particularly in older adults, there is a crucial need for practical and scalable tools to monitor and manage VTE risk in oncology settings.17 In this context, further investigation into whether IC could serve as a VTE risk assessment tool for patients with cancer holds practical value.
The heritability of VTE has been estimated to be as high as 50%, indicating a strong genetic predisposition to this disease.18 Established polygenic risk scores (PRS) can be used to assess an individual’s genetic susceptibility to VTE.19 Previous studies have demonstrated that disease susceptibility can be modified by individual conditions (such as accelerated aging or environmental pollution).20 However, it remains unclear whether IC and genetic predisposition exert a synergistic effect on the onset and progression of VTE. To address these existing gaps, we used and combined information from a large-scale, population-based database, the UK Biobank, to investigate the association of IC status and incident VTE in both the general population and patients with cancer.21
Methods
Study participants
The UK Biobank study is a population-based retrospective cohort study, recruiting more than 500,000 UK residents aged 40-69 years from 2006 to 2010. At enrollment, detailed information on social-demographics, behavioral factors, physical and cognitive functions, environmental exposures, lifestyle and medical history were collected. Biological samples including blood, urine, and saliva were also obtained, from which genetic data were determined. Details of the UK Biobank cohort are reported elsewhere.22 Our study was reviewed and approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Multicenter Research Ethics Committee (11/NW/0382), and data used in our study was accessed through the Biobank consortium (Application No: 100739). Our study was in accordance with the Declaration of Helsinki with informed consent from all participants.
In the current study, we included participants with available data on IC, defined by Beard et al5 and exclude those (1) with a history of VTE or stroke or heart failure or coronary heart disease or myocardial infarction or Parkinson disease; (2) with a body mass index (BMI) lower than 18.5 kg/m2. The detailed inclusion and exclusion process is shown in Supplementary Fig 1 (online only). An outline of the study is depicted in Fig 1.
Fig 1.
Outline of the study.
Exposure
In the current study, functional ability was defined by IC score, which was assessed by five components encompassing psychological capacity, sensory capacity, vitality capacity, locomotor capacity, and cognitive capacity. The psychological domain included self-reported exhaustion and sleep duration. The sensory domain included eye problems and hearing difficulty. The vitality domain included measured grip strength and self-reported weight loss. The locomotor domain was assessed by self-reported gait speed (see details in Supplementary Table I, online only). We did not measure the cognitive capacity because this was not included in the UK Biobank assessment. An IC score was the sum of the scores obtained from the seven factors in the four domains mentioned above. Based on these criteria, participants were categorized into 0 factors (none of the criteria met), 1 factor (1 criterion met), 2 factors (2 criteria met), 3 factors (3 criteria met), and ≥4 factors (4 or more criteria met), provided there were no missing data on any of the four domains.
Outcomes
The primary disease outcome was VTE. Diagnosis of VTE was determined by medical records linked to the National Health Service Digital with the following International Classification of Diseases, Ninth Revision) codes, ICD-10 codes, and Office of Population Censuses and Surveys Classification of Surgical Operations and Procedures, version 4 codes (see details in Supplementary Table II, online only), which incorporated primary care, hospital admissions, self-reports, death records, and results from verbal interviews or touchscreen questionnaires. The follow-up period was defined as the time between baseline visit to the first occurrence of VTE or lost to follow-up or death or censoring date (January 1, 2024), whichever came the earliest.
We further divided the participants with incident VTE into two groups, according to whether the incident of cancer occurred before the diagnosis of VTE.
Covariates
Factors associated with VTE were included as confounding factors in the analyses, including chronological age, sex, ethnicity (White/others), educational level (colleges or above/below colleges), Townsend deprivation index (TDI), BMI, systolic blood pressure, glycated hemoglobin, high-density lipoprotein, smoking status (never/former/current), drinking status (never/former/current), healthy diet (yes/no), meeting physical activity recommendation (yes/no) defined according to the UK guidelines of 150 minutes of walking or moderate activity per week or 75 minutes of vigorous activity,23 use of statins (yes/no), use of anticoagulant drugs (yes/no), and use of antiplatelet drugs (yes/no). TDI represented socioeconomic status level, with lower value indicates higher the socioeconomic status. Healthy diet was classified as yes if participants consumed macronutrients in appropriate proportions to support energetic and physiologic needs without excessive intake, as well as providing adequate micronutrients and hydration. Genetic data of VTE was based on the standard PRS (field ID 262112 in the UK Biobank) supported by the external genome-wide association studies data. Charlson comorbidity index was used to define the status of comorbidity.24
Statistical analysis
The baseline characteristics of all participants were presented as mean [standard deviation (SD) or median [interquartile range] for continuous variables and number (%) for categorical variables. Comparisons between different groups at baseline were conducted with analysis of variance and χ2 tests. Cox proportional hazards regression models were fitted to investigate the associations between IC score and future risk of incident VTE/VTE occurred before or after cancer with multiple confounding factors adjusted. Two major models were applied with model I adjusted for age, sex, and ethnicity; model II further adjusted for educational level, TDI, BMI, glycated hemoglobin, high-density lipoprotein, smoking status, drinking status, healthy diet, physical activity, medication with statins, anticoagulant drugs, and antiplatelet drugs. Restricted cubic spline models were used to depict the associations between the number of IC scores with the risk of VTE. To examine the joint association between IC score and VTE-PRS, we further classified the participants into groups with low, intermediate and high genetic risk according to the tertiles of PRS and treated participants with low-PRS and 0 IC factors as the reference group. The interaction analysis between IC status and categories of PRS to VTE was performed by using the likelihood ratio test comparing models with and without a cross-product term.
Sensitivity analyses were further performed to validate the findings. First, we further adjusting for the competing risk of death in the cox regression models. Second, we excluded those diagnosed with VTE within 1 year of enrollment to minimize the effect of reverse causality. Third, we further adjusted for the history of any fracture. Fourth, we further adjusting for the Charlson comorbidity index. All statistical analyses were performed with Stata (version 18, StataCorp) and R (version 4.4.3, R Foundation for Statistical Computing, www.R-project.org). A two-sided P value of < .05 was considered as statistically significant.
Results
Baseline characteristics
Of the 502,357 participants, 415,581 met the eligibility criteria (median age, 57.0 years; interquartile range, 50.0-63.0; 54.96% female; 94.89% White). During a median follow-up of 14.37 years (range, 0.008-17.1), 8299 (2.00%) developed VTE, including 3878 (0.94%) with and 4421 (1.07%) without a history of cancer. Compared with those without incident of VTE, patients who developed VTE were more likely to be older, male and had higher IC score (Table I). At baseline, 155,300 (37.37%) had 0 IC factors, 153,619 (36.96%) had 1, 75,140 (18.08%) had 2, 24,444 (5.88%) had 3, and 7078 (1.70%) had ≥4 factors. Participants with higher IC score tended to be older, have higher BMI, and take more medications. The detailed baseline characteristics are displayed in Table I, Supplementary Tables III and IV (online only).
Table I.
Baseline characteristics of all participants and stratified by incident venous thromboembolism (VTE)
| Variables | All VTE (N = 415,581) | Nonincident VTE (N = 407,282) | Incident VTE (N = 8299) | P value |
|---|---|---|---|---|
| Age, years, median [IQR] | 57 [50, 63] | 57 [50, 63] | 61 [55, 65] | <.001 |
| Female | 228,391 (54.96) | 224,353 (55.09) | 4038 (48.66) | <.001 |
| Ethnicity | – | – | – | <.001 |
| White | 394,342 (94.89) | 386,359 (94.86) | 7983 (96.19) | – |
| Others | 21,239 (5.11) | 20,923 (5.14) | 316 (3.81) | |
| Education level | – | – | – | <.001 |
| Colleges or above | 49,219 (11.84) | 48,401 (11.88) | 818 (9.86) | – |
| Below colleges | 366,362 (88.16) | 358,881 (88.12) | 7481 (90.14) | |
| Townsend deprivation index, median (IQR) | −2.24 (−3.70, 0.32) | -2.24 (−3.70, 0.31) | −2.02 (−3.57, 0.76) | <.001 |
| BMI, kg/cm2, mean (SD) | 27.30 (4.65) | 27.27 (4.64) | 28.80 (5.18) | <.001 |
| SBP, mmHg, mean (SD) | 137.77 (18.58) | 137.72 (18.58) | 140.22 (18.57) | <.001 |
| HbA1c, mean (SD) | 35.82 (6.33) | 35.80 (6.30) | 37.00 (7.50) | <.001 |
| HDL, mean (SD) | 1.46 (0.38) | 1.46 (0.38) | 1.41 (0.38) | <.001 |
| Smoking status | – | – | – | <.001 |
| Never | 230,898 (55.72) | 226,880 (55.86) | 4018 (48.62) | – |
| Former | 141,429 (34.13) | 138,220 (34.03) | 3209 (38.83) | |
| Current | 42,074 (10.15) | 41,037 (10.10) | 1037 (12.55) | |
| Drinking status | – | – | – | <.001 |
| Never | 16,680 (4.02) | 16,317 (4.01) | 363 (4.38) | – |
| Former | 13,435 (3.24) | 13,073 (3.21) | 362 (4.37) | |
| Current | 385,180 (92.75) | 377,620 (92.78) | 7560 (91.25) | |
| Healthy diet | 31,527 (7.59) | 30,980 (7.61) | 547 (6.59) | .001 |
| Meeting PA recommendation | 272,639 (82.24) | 267,422 (82.24) | 5217 (81.78) | .340 |
| Use of statins | 24,831 (5.98) | 24,189 (5.94) | 642 (7.74) | <.001 |
| Use of anticoagulant drugs | 3321 (0.80) | 3229 (0.79) | 92 (1.11) | .001 |
| Use of antiplatelet drugs | 45,358 (10.91) | 44,104 (10.83) | 1254 (15.11) | <.001 |
| Intrinsic capacity scoring | – | – | – | <.001 |
| 0 factors | 155,300 (37.37) | 152,759 (37.51) | 2541 (30.62) | – |
| 1 factor | 153,619 (36.96) | 150,581 (36.97) | 3038 (36.61) | |
| 2 factors | 75,140 (18.08) | 73,395 (18.02) | 1745 (21.03) | |
| 3 factors | 24,444 (5.88) | 23,734 (5.83) | 710 (8.56) | |
| ≥4 factors | 7078 (1.70) | 6813 (1.67) | 265 (3.19) | |
| Genetic risk category | – | – | – | <.001 |
| Low | 135,335 (33.33) | 133,581 (33.57) | 1754 (21.69) | – |
| Intermediate | 135,334 (33.33) | 132,867 (33.39) | 2467 (30.51) | |
| High | 135,334 (33.33) | 131,469 (33.04) | 3865 (47.80) |
BMI, Body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; IQR, interquartile range; PA, physical activity; SBP, systolic blood pressure; VTE, venous thromboembolism.
Data presented as n (%) or median (interquartile range) as appropriate.
Association between IC and future risk of VTE
As shown in Fig 2, the risk of VTE increased with higher IC scores (P-linearity < .001, P-nonlinearity < .001). This association was consistent among individuals with or without a history of cancer. In multivariable Cox regression models, compared to those with 0 IC factors, participants with 1 (HR, 1.08; 95% CI, 1.01-1.16; P = .029), 2 (HR, 1.12; 95% CI, 1.04-1.22; P = .005), 3 (HR, 1.19; 95% CI, 1.06-1.34; P = .003), and ≥4 IC factors (HR, 1.46; 95% CI, 1.23-1.75; P < .001) had significantly increased VTE risk (Model 2, Table II).
Fig 2.
Restricted cubic spline plots showing the associations between intrinsic capacity scores with the risk of venous thromboembolism (VTE) in all participants (A), patients with cancer (B), and noncancer individuals (C).
Table II.
Association between intrinsic capacity and future risk of venous thromboembolism (VTE)
| Intrinsic capacity score | HR (95% CI) |
|||||
|---|---|---|---|---|---|---|
| All VTE | P value | VTE with history of cancer | P value | VTE without history of cancer | P value | |
| Model 1a | ||||||
| 0 factors | 1 | Reference | 1 | Reference | 1 | Reference |
| 1 factor | 1.13 (1.07, 1.19) | <.001 | 1.10 (1.02, 1.18) | .019 | 1.15 (1.07, 1.24) | <.001 |
| 2 factors | 1.28 (1.21, 1.36) | <.001 | 1.20 (1.09, 1.31) | <.001 | 1.36 (1.25, 1.48) | <.001 |
| 3 factors | 1.61 (1.48, 1.75) | <.001 | 1.44 (1.27, 1.63) | <.001 | 1.77 (1.58, 1.98) | <.001 |
| ≥4 factors | 2.07 (1.83, 2.36) | <.001 | 1.87 (1.54, 2.26) | <.001 | 2.27 (1.91, 2.69) | <.001 |
| Model 2b | ||||||
| 0 factors | 1 | Reference | 1 | Reference | 1 | Reference |
| 1 factor | 1.08 (1.01, 1.16) | .029 | 1.03 (0.93, 1.13) | .617 | 1.13 (1.03, 1.24) | .012 |
| 2 factors | 1.12 (1.04, 1.22) | .005 | 1.06 (0.94, 1.19) | .374 | 1.19 (1.06, 1.32) | .002 |
| 3 factors | 1.19 (1.06, 1.34) | .003 | 1.14 (0.96, 1.36) | .134 | 1.24 (1.06, 1.46) | .007 |
| ≥4 factors | 1.46 (1.23, 1.75) | <.001 | 1.35 (1.03, 1.76) | .031 | 1.57 (1.24, 1.98) | <.001 |
CI, Confidence interval; HR, hazard ratio.
Model 1: Adjusted for age, sex, and ethnicity.
Model 2: Further adjusted for educational level, Townsend deprivation index, body mass index, glycated hemoglobin, high-density lipoprotein, smoking status, drinking status, healthy diet, physical activity, medication of statins, and anticoagulant and antiplatelet drugs.
Among participants with a prior cancer diagnosis, those with ≥4 IC factors remained at higher risk (HR, 1.35; 95% CI, 1.03-1.76; P = .031), whereas associations for lower IC scores were not statistically significant. In participants without cancer, higher IC scores were consistently associated with increased VTE risk: 1 factor (HR, 1.13; 95% CI, 1.03-1.24; P = .012), 2 (HR, 1.19; 95% CI, 1.06-1.32; P = .002), 3 (HR, 1.24, 95% CI, 1.06-1.46; P = .007), and ≥4 factors (HR, 1.57; 95% CI, 1.24-1.78; P < .001) (Model 2, Table II).
Association between IC, PRS and future risk of VTE
Across all participants, higher IC scores were associated with greater VTE risk regardless of genetic predisposition. Individuals with high PRS and ≥4 IC factors had the highest VTE risk compared to those with low PRS and no IC factors (HR, 2.99; 95% CI, 2.29-3.90; P < .001; interaction P < .001; Fig 3). Individuals with low genetic risk but ≥4 IC factors had a comparable VTE risk to those with high genetic risk and no IC factors (HR, 2.17; 95% CI, 1.91-2.47; vs HR, 1.86; 95% CI, 1.32-2.60; P > .05). Subgroup analysis showed that the association between higher IC scores and incident VTE remained consistent in individuals with or without a history of cancer (Supplementary Fig 2, online only).
Fig 3.
Association between intrinsic capacity, genetic risk, and risk of incident venous thromboembolism (VTE). CI, Confidence interval; HR, hazard ratio; PRS, polygenic risk score.
Compared with participants with the worst IC status (≥4 factors), those with optimal IC (0 factors) had significantly lower VTE risk across genetic risk categories: low PRS (HR, 0.52; 95% CI, 0.37-0.73; P < .001), high PRS (HR, 0.73; 95% CI, 0.56-0.94; P = .015), though not significant for intermediate PRS (HR, 0.80; 95% CI, 0.56-1.13; P = .201; Supplementary Table V, online only).
Sensitivity analyses
Sensitivity analyses found that the association between IC and all VTE and participants with or without cancer before VTE remained stable after adjusting for all-cause mortality as competing risk (Supplementary Table VI, online only); excluding incident VTE in 1-year after baseline IC assessment (Supplementary Table VII, online only); adjusting for history of fracture (Supplementary Table VIII, online only); and adjusting for Charlson comorbidity index (Supplementary Table IX, online only).
Discussion
This large-scale population-based study using data from the UK Biobank found that a worse IC status, as indicated by a higher IC score, was significantly associated with an increased risk of VTE, independent of sociodemographic and lifestyle factors, in both the general population and individuals with cancer. Notably, individuals with an optimal IC status (0 IC score) had the lowest risk of VTE, underscoring the potential protective role of preserved IC. Furthermore, the association between IC and VTE was particularly pronounced among participants with a high PRS for VTE.
Although IC has been recognized as a predictor of cardiovascular diseases, its role in thrombotic disorders such as VTE has not been systematically investigated. To the best of our knowledge, this is the first study to demonstrate an association between IC and VTE, and to explore its interaction with genetic predisposition. The observation that IC-related VTE risk was amplified in individuals with high genetic risk suggests a synergistic effect between biological ageing and inherited susceptibility. Similar gene-environment interactions have been observed in other age-related conditions; for example, Sun et al9 reported a stronger association between IC impairment and dementia risk among individuals with elevated genetic risk for dementia. Although genetic testing can identify individuals at elevated VTE risk, its use in clinical practice is limited by cost and complexity.2,25 In contrast, IC assessment, based on low-cost, noninvasive measures, provides a feasible tool for early-risk stratification. Our findings suggest that incorporating IC screening may be particularly beneficial in individuals with high genetic risk, offering a more holistic and personalized approach to VTE prevention in ageing populations.
Notably, we observed that poorer IC status was significantly associated with higher VTE risk among individuals with a history of cancer. This aligns with the well-established concept that cancer and its treatments promote hypercoagulability through multiple biological pathways.15,16 Although VTE is a leading cause of mortality in patients with cancer, few clinically practical predictors are available for identifying those at high risk. Our findings suggest that IC assessment could be considered not only during hospitalization or perioperative care, but also in asymptomatic phases of cancer. Future studies should validate the predictive value of IC across cancer types and disease stages, to better inform early prophylaxis, targeted diagnostic test and tailored anticoagulant treatment.
Previous studies have shown that worse IC status is associated with elevated inflammatory markers, including C-reactive protein, interleukin-6, and tumor necrosis factor receptor 1, suggesting chronic low-grade inflammation as a potential mechanism.26, 27, 28, 29 Inflammation can trigger systemic coagulation dysfunction by promoting immune thrombosis and enhancing platelet-neutrophil interactions, which in turn damage the endothelium and disrupt normal hemostasis.30 In parallel, poor IC is linked to metabolic disturbances such as hyperglycemia, which further contribute to thrombosis via impaired fibrinolysis, increased platelet reactivity, and activation of the coagulation cascade.7 Hyperglycemia also directly injures the vascular endothelium, increases vascular permeability, and induces the expression of adhesion molecules such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1.31, 32, 33, 34, 35 Together, these biological processes, chronic inflammation, metabolic dysfunction, and endothelial injury, may underlie the observed link between impaired IC and elevated VTE risk, in line with Virchow triad of hypercoagulability, stasis, and endothelial dysfunction.36
In 2017, the WHO released the Integrated Care for Older People (ICOPE) guidelines to address declines in IC,37 followed by the 2019 Handbook on Person-Centred Assessment and Pathways in Primary Care,38 in which IC screening was proposed as the first step of the ICOPE approach. A personalized care plan can be developed to support healthy ageing based on simple IC measurements, which is feasible for use in community, primary care, and follow-up settings by trained health care and social workers.39 Our study further reinforces the clinical utility and necessity of IC screening in the elderly. Given its noninvasive nature, low cost, and rapid implementation, IC assessment holds great promise as a feasible tool in both community-based screening and hospital settings. Identifying individuals with impaired IC may facilitate earlier and more personalized VTE prevention strategies, including intensified monitoring and timely prophylaxis. Moreover, precise IC stratification enables more efficient allocation of health care resources to vulnerable populations, potentially reducing unnecessary health care expenditures. The WHO-ICOPE application, freely available on both iOS and Android platforms,40,41 further supports the integration of IC self-monitoring into routine health management.
It is noteworthy that individuals with stable or improved IC have been shown to exhibit significantly lower risks of all-cause and cause-specific mortality, including cardiovascular and organ-related deaths, as well as reduced morbidity and hospitalization rates.42 A range of interventions, such as exercise training,43 healthy dietary habits,44 and cognitive stimulation therapy,45 have proven effective in maintaining or enhancing IC. These findings support the notion that IC is a modifiable risk factor for VTE, and its improvement may offer protective benefits. Currently, the prevention of VTE relies primarily on pharmacologic anticoagulation, yet its associated bleeding risk limits its use among the majority of the senior population.46 This underscores an urgent need for safer nonpharmacological preventive strategies. These findings offer a potential promising solution for nonpharmacological VTE prevention. Improving IC through multidimensional interventions mentioned above may serve as an important complement to current VTE preventive strategies, thereby further reducing the overall disease burden and health care costs among the elderly. Longitudinal studies are warranted to confirm the impact of IC-targeted interventions on VTE prevention. In addition, frailty, as another well-known ageing indicator, has been shown in previous studies to be closely associated with VTE.19 Given that frailty reflects an individual’s accumulation of deficits, while IC reflects an individual's functional reserve, the two indicators may have complementary efficacy in predicting VTE. Therefore, future studies may consider combining IC, frailty, and genetic risk to achieve more accurate VTE risk prediction.
This study possessed several strengths including the population-based cohort design with a large sample size and a long follow-up period. In addition, comprehensive confounding factors were adjusted and several sensitivity analyses were performed to ensure the robustness of our findings. However, several limitations should be noted. First, the selection bias of the UK Biobank cohort cannot be overlooked since most of the participants are of White ethnicity, relatively healthy and younger, and more highly educated than the general population, which may preclude the direct extrapolation to other ethnicities and populations.47 Future studies involving multiethnicities are warranted to validate the findings. Second, the assessment of IC was partly based on self-reported information, which might lead to recall and misclassification bias. Nonetheless, the currently used metrics for IC are commonly acknowledged and proposed in the ICOPE guideline. Third, residual confounding still existed even after adjusting for multiple variables. Fourthly, we were unable to depict the longitudinal trajectory of IC status due to the lack of multi-timepoint assessments. Moreover, we could not include the complete cognitive capacity subdomain in our analysis because certain information was not available in the UK Biobank. In addition, our VTE definition relied on clinical diagnosis codes from linked electronic health records without a requirement for imaging confirmation. Accordingly, some clinically silent or atypical events that did not lead to health care contact may have been missed, likely underestimating the true VTE incidence. However, our approach is consistent with previous UK Biobank studies,48, 49, 50 making substantial misclassification of recorded events unlikely.
Conclusion
In conclusion, this large-scale study demonstrated that individuals with worse IC status were at increased risk of incident VTE in both the general population and patients with cancer. This relationship was strengthened in patients with higher genetic risk of VTE. Collectively, our findings suggest the feasibility of IC measurement in VTE risk assessment. Future research is warranted to investigate the predictive value of IC and the effect of longitudinal trajectories of IC on VTE.
Author Contributions
YC and JX contributed equally to this article and share co-first authorship. YS made substantial contributions to this study. KH serves as the post-publication corresponding author for this study.
Conception and design: YC, KH, YS
Analysis and interpretation: YC, JX, RS, KH, YS
Data collection: YC, JX, RS, KH, YS
Writing the article: YC, JX, RS, KH, YS
Critical revision of the article: YC, JX, RS, KH, YS
Final approval of the article: YC, JX, RS, KH, YS
Statistical analysis: YC
Obtained funding: KH, YS
Overall responsibility: KH, YS
Data access and sharing statements
Data supporting the findings of this study are available at https://ams.ukbiobank.ac.uk/. As these data were used under the current research license, they are not publicly available. However, upon reasonable request and with the permission of the UK Biobank, data can be obtained from Kai Huang (huangk37@mail.sysu.edu.cn).
Funding
This study was supported by the National Natural Science Foundation of China Grant 81800420 and Basic and Applied Basic Research Projects of Guangzhou Basic Research Plan (2023A04J2116).
Disclosures
None.
Footnotes
The editors and reviewers of this article have no relevant financial relationships to disclose per the Journal policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest.
Additional material for this article may be found online at www.jvsvenous.org.
Appendix
Additional material for this article may be found online at www.jvsvenous.org.
Appendix (online only)
References
- 1.Fernandes C.J., Morinaga L.T.K., Alves J.L., et al. Cancer-associated thrombosis: the when, how and why. Eur Respir Rev. 2019;28 doi: 10.1183/16000617.0119-2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Rosendaal F.R. Venous thrombosis: a multicausal disease. Lancet. 1999;353:1167–1173. doi: 10.1016/s0140-6736(98)10266-0. [DOI] [PubMed] [Google Scholar]
- 3.World Health O . World Health Organization; 2015. World report on ageing and health. [Google Scholar]
- 4.Belloni G., Cesari M. Frailty and intrinsic capacity: two distinct but related constructs. Front Med. 2019;6:133. doi: 10.3389/fmed.2019.00133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Beard J.R., Officer A., de Carvalho I.A., 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]
- 6.Sánchez-Sánchez J.L., Lu W.-H., Gallardo-Gómez D., et al. Association of intrinsic capacity with functional decline and mortality in older adults: a systematic review and meta-analysis of longitudinal studies. Lancet Healthy Longevity. 2024;5:e480–e492. doi: 10.1016/S2666-7568(24)00092-8. [DOI] [PubMed] [Google Scholar]
- 7.Lee W.J., Peng L.N., Lin M.H., Loh C.H., Hsiao F.Y., Chen L.K. Intrinsic capacity differs from functional ability in predicting 10-year mortality and biological features in healthy aging: results from the I-Lan longitudinal aging study. Aging (Albany NY) 2023;15:748–764. doi: 10.18632/aging.204508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ramírez-Vélez R., Iriarte-Fernández M., Santafé G., et al. 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]
- 9.Sun M., He Q., Sun N., et al. Intrinsic capacity, polygenic risk score, APOE genotype, and risk of dementia: a prospective cohort study based on the UK biobank. Neurology. 2024;102 doi: 10.1212/WNL.0000000000209452. [DOI] [PubMed] [Google Scholar]
- 10.Zhu L., Zong X., Shi X., Ouyang X. Association between intrinsic capacity and Sarcopenia in hospitalized older patients. J Nutr Health Aging. 2023;27:542–549. doi: 10.1007/s12603-023-1946-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Stolz E., Mayerl H., Freidl W., Roller-Wirnsberger R., Gill T.M. Intrinsic capacity predicts negative health outcomes in older adults. J Gerontol A Biol Sci Med Sci. 2022;77:101–105. doi: 10.1093/gerona/glab279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zeng X., Shen S., Xu L., et al. The impact of intrinsic capacity on adverse outcomes in older hospitalized patients: a one-year Follow-Up study. Gerontology. 2021;67:267–275. doi: 10.1159/000512794. [DOI] [PubMed] [Google Scholar]
- 13.Anderson F.A., Jr., Spencer F.A. Risk factors for venous thromboembolism. Circulation. 2003;107(23 Suppl 1):I9–I16. doi: 10.1161/01.CIR.0000078469.07362.E6. [DOI] [PubMed] [Google Scholar]
- 14.Barreto P.S. Intrinsic capacity and aging: advances in research and clinical practice. J Nutr Health Aging. 2024;28 doi: 10.1016/j.jnha.2024.100336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mulder F.I., Horváth-Puhó E., van Es N., et al. Venous thromboembolism in cancer patients: a population-based cohort study. Blood. 2021;137:1959–1969. doi: 10.1182/blood.2020007338. [DOI] [PubMed] [Google Scholar]
- 16.Guntupalli S.R., Spinosa D., Wethington S., Eskander R., Khorana A.A. Prevention of venous thromboembolism in patients with cancer. BMJ. 2023;381 doi: 10.1136/bmj-2022-072715. [DOI] [PubMed] [Google Scholar]
- 17.Steinmeyer Z., Berbon C., Sourdet S., Gérard S., Rolland Y., Balardy L. Letter to the the editor: the WHO ICOPE program to monitor intrinsic capacity in older adults with cancer. J Frailty Aging. 2024;13:71–72. doi: 10.14283/jfa.2024.6. [DOI] [PubMed] [Google Scholar]
- 18.Ghouse J., Tragante V., Ahlberg G., et al. Genome-wide meta-analysis identifies 93 risk loci and enables risk prediction equivalent to monogenic forms of venous thromboembolism. Nat Genet. 2023;55:399–409. doi: 10.1038/s41588-022-01286-7. [DOI] [PubMed] [Google Scholar]
- 19.Xu J., Ren Q., Su Y., Lin L., Shen R., Huang K. Frailty, genetic risk, and long-term risk of venous thromboembolism: insight from a UK biobank study. Am J Hematol. 2025;100:1523–1532. doi: 10.1002/ajh.27753. [DOI] [PubMed] [Google Scholar]
- 20.Wang L., Xie J., Hu Y., Tian Y. Air pollution and risk of chronic obstructed pulmonary disease: the modifying effect of genetic susceptibility and lifestyle. EBioMedicine. 2022;79 doi: 10.1016/j.ebiom.2022.103994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Khorana A.A., Mackman N., Falanga A., et al. Cancer-associated venous thromboembolism. Nat Rev Dis Primers. 2022;8:11. doi: 10.1038/s41572-022-00336-y. [DOI] [PubMed] [Google Scholar]
- 22.Sudlow C., Gallacher J., Allen N., et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12 doi: 10.1371/journal.pmed.1001779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Department of Health & Social Care WG, Department of Health Northern Ireland & Scottish Government . Care DoHS. Department of Health & Social Care; 2019. UK Chief Medical Officers’ PhysicalActivity Guidelines. [Google Scholar]
- 24.Charlson M.E., Pompei P., Ales K.L., MacKenzie C.R. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 25.Zöller B., Svensson P.J., Dahlbäck B., Lind-Hallden C., Hallden C., Elf J. Genetic risk factors for venous thromboembolism. Expert Rev Hematol. 2020;13:971–981. doi: 10.1080/17474086.2020.1804354. [DOI] [PubMed] [Google Scholar]
- 26.Giudici K.V., de Souto Barreto P., Guerville F., et al. Associations of C-reactive protein and homocysteine concentrations with the impairment of intrinsic capacity domains over a 5-year follow-up among community-dwelling older adults at risk of cognitive decline (MAPT Study) Exp Gerontol. 2019;127 doi: 10.1016/j.exger.2019.110716. [DOI] [PubMed] [Google Scholar]
- 27.Beyene M.B., Visvanathan R., Amare A.T. Intrinsic capacity and its biological basis: a scoping review. J Frailty Aging. 2024;13:193–202. doi: 10.14283/jfa.2024.30. [DOI] [PubMed] [Google Scholar]
- 28.Lu W.H., Gonzalez-Bautista E., Guyonnet S., et al. 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]
- 29.Ma L., Liu P., Zhang Y., Sha G., Zhang L., Li Y. High serum tumor necrosis factor receptor 1 levels are related to risk of low intrinsic capacity in elderly adults. J Nutr Health Aging. 2021;25:416–418. doi: 10.1007/s12603-020-1533-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kubes P. The versatile platelet contributes to inflammation, infection, hemostasis, coagulation and cancer. Semin Immunol. 2016;28:535. doi: 10.1016/j.smim.2016.11.002. [DOI] [PubMed] [Google Scholar]
- 31.Panchagnula N., Brasher W.P. Hyperglycemia and venous thromboembolism. Diagnostics (Basel) 2024;14:1994. doi: 10.3390/diagnostics14171994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Bryk-Wiązania A.H., Undas A. Hypofibrinolysis in type 2 diabetes and its clinical implications: from mechanisms to pharmacological modulation. Cardiovasc Diabetol. 2021;20:191. doi: 10.1186/s12933-021-01372-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Shilpi K., Potekar R.M. A Study of platelet indices in type 2 diabetes mellitus patients. Indian J Hematol Blood Transfus. 2018;34:115–120. doi: 10.1007/s12288-017-0825-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Amalia M., Puteri M.U., Saputri F.C., Sauriasari R., Widyantoro B. Platelet Glycoprotein-Ib (GPIb) May serve as a Bridge between type 2 diabetes mellitus (T2DM) and atherosclerosis, making it a potential target for antiplatelet agents in T2DM patients. Life (Basel) 2023;13:1473. doi: 10.3390/life13071473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lemkes B.A., Hermanides J., Devries J.H., Holleman F., Meijers J.C., Hoekstra J.B. Hyperglycemia: a prothrombotic factor? J Thromb Haemost. 2010;8:1663–1669. doi: 10.1111/j.1538-7836.2010.03910.x. [DOI] [PubMed] [Google Scholar]
- 36.Bagot C.N., Arya R. Virchow and his triad: a question of attribution. Br J Haematol. 2008;143:180–190. doi: 10.1111/j.1365-2141.2008.07323.x. [DOI] [PubMed] [Google Scholar]
- 37.World Health O . World Health Organization; 2017. Integrated care for older people (ICOPE): guidelines on community-level interventions to manage declines in intrinsic capacity. Report No.: Contract No.: WHO/MCA/17.06.10. [PubMed] [Google Scholar]
- 38.World Health O . World Health Organization; 2019. Integrated care for older people (ICOPE): guidance for person-centred assessment and pathways in primary care. Report No.: Contract No.: WHO/FWC/ALC/19.1. [Google Scholar]
- 39.Takeda C., Guyonnet S., Sumi Y., Vellas B., Araujo de Carvalho I. Integrated care for older people and the implementation in the INSPIRE care cohort. J Prev Alzheimers Dis. 2020;7:70–74. doi: 10.14283/jpad.2020.8. [DOI] [PubMed] [Google Scholar]
- 40.Sanchez-Rodriguez D., Annweiler C., Gillain S., Vellas B. Implementation of the integrated care of older people (ICOPE) app in primary care: new technologies in geriatric care during quarantine of COVID-19 and beyond. J Frailty Aging. 2021;10:139–140. doi: 10.14283/jfa.2020.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Sanchez-Rodriguez D., Piccard S., Dardenne N., Giet D., Annweiler C., Gillain S. Implementation of the integrated care of older people (ICOPE) app and ICOPE monitor in primary care: a study protocol. J Frailty Aging. 2021;10:290–296. doi: 10.14283/jfa.2021.22. [DOI] [PubMed] [Google Scholar]
- 42.Hwang A.C., Chen L.Y., Tseng S.H., et al. Intrinsic capacity transitions predict overall and cause-specific mortality, incident disability, and healthcare utilization. J Nutr Health Aging. 2024;28 doi: 10.1016/j.jnha.2024.100359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Jones R., Taylor T.L., Mankowski R.T., et al. Exercise training to preserve vitality capacity in ageing. Exp Physiol. 2026;111:2805–2812. doi: 10.1113/EP091731. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Lim K.Y., Lo H.C., Cheong I.F., et al. Healthy eating enhances intrinsic capacity, thus promoting functional ability of retirement home residents in Northern Taiwan. Nutrients. 2022;14:2225. doi: 10.3390/nu14112225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Merchant R.A., Chan Y.H., Anbarasan D., Woo J. The impact of exercise and cognitive stimulation therapy on intrinsic capacity composite score in pre-frail older adults: a pre-post intervention study. J Frailty Aging. 2024;13:131–138. doi: 10.14283/jfa.2024.16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Tritschler T., Kraaijpoel N., Le Gal G., Wells P.S. Venous thromboembolism: advances in diagnosis and treatment. JAMA. 2018;320:1583–1594. doi: 10.1001/jama.2018.14346. [DOI] [PubMed] [Google Scholar]
- 47.Fry A., Littlejohns T.J., Sudlow C., et al. Comparison of sociodemographic and health-related characteristics of UK biobank participants with those of the general population. Am J Epidemiol. 2017;186:1026–1034. doi: 10.1093/aje/kwx246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Guman N.A.M., Mulder F.I., Ferwerda B., et al. Polygenic risk scores for prediction of cancer-associated venous thromboembolism in the UK Biobank cohort study. J Thromb Haemost. 2023;21:3175–3183. doi: 10.1016/j.jtha.2023.07.009. [DOI] [PubMed] [Google Scholar]
- 49.Yuan S., Sun Y., Chen J., Li X., Larsson S.C. Long-term risk of venous thromboembolism among patients with gastrointestinal non-neoplastic and neoplastic diseases: a prospective cohort study of 484 211 individuals. Am J Hematol. 2024;99:172–181. doi: 10.1002/ajh.27106. [DOI] [PubMed] [Google Scholar]
- 50.Kolin D.A., Kulm S., Elemento O. Prediction of primary venous thromboembolism based on clinical and genetic factors within the U.K. Biobank. Sci Rep. 2021;11 doi: 10.1038/s41598-021-00796-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.