[A prospective dual-cohort study on the association between frailty index outcomes and cardiovascular disease onset risk in elderly diabetic populations].
Li Y, Li S, Jin A • Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine] • 2026
Deterioration to or persistence of frailty is associated with an elevated risk of CVD in older adults with diabetes, and these associations remained robust after competing-risk validation.
Key Findings
Results
The 'pre-frail→frail' transition group had significantly higher risk of incident CVD compared to the 'robust→robust' group.
Pooled HR from two-cohort meta-analysis: HR=2.04 (95%CI: 1.66–2.51), P<0.001
After accounting for competing risks (non-CVD death), subdistribution HR (sHR)=1.90 (95%CI: 1.55–2.33)
Results were consistent across both CHARLS and ELSA cohorts
Cox proportional hazards models and Fine-Gray subdistribution hazard models were used
Results
The 'frail→frail' (persistent frailty) group had significantly higher risk of incident CVD compared to the 'robust→robust' group.
Pooled HR from two-cohort meta-analysis: HR=2.03 (95%CI: 1.63–2.53), P<0.001
After competing-risk adjustment, sHR=1.87 (95%CI: 1.50–2.31)
This group represented individuals who remained frail across both measurement time points
Fixed-effects models were used to meta-analyze results from both cohorts
Results
The 'robust→pre-frail/frail' transition group had significantly elevated risk of incident CVD compared to the 'robust→robust' group.
Pooled HR from two-cohort meta-analysis: HR=1.60 (95%CI: 1.32–1.94), P<0.001
After competing-risk adjustment, sHR=1.50 (95%CI: 1.23–1.83)
This group represents individuals who were initially robust but deteriorated to pre-frail or frail status
Magnitude of risk was lower than persistent frailty or pre-frail→frail transition groups
Methods
The study enrolled diabetic older adults from two longitudinal cohorts with substantial numbers of incident CVD cases documented during follow-up.
891 diabetic participants enrolled from the China Health and Retirement Longitudinal Study (CHARLS) with 439 incident CVD cases
827 diabetic participants enrolled from the English Longitudinal Study of Ageing (ELSA) with 344 incident CVD cases
Participants were classified into seven frailty index (FI) transition patterns
Study design was a prospective dual-cohort study
Results
Competing-risk analysis treating non-CVD death as the competing event confirmed the associations between frailty transitions and CVD incidence remained robust.
Fine-Gray subdistribution hazard models were applied to account for competing risks
All three high-risk transition groups retained statistically significant elevated CVD risk after competing-risk adjustment
The sHR estimates were modestly attenuated compared to standard Cox HR estimates but remained clinically and statistically meaningful
Pooled meta-analysis using fixed-effects models was conducted across both cohorts for competing-risk results
What This Means
This research suggests that how frailty changes over time in older adults with diabetes has a strong relationship with their risk of developing cardiovascular disease (CVD). The study followed nearly 1,800 older diabetic adults across two large longitudinal studies—one in China and one in England—and tracked how their frailty status shifted over time. Participants were grouped into seven patterns based on whether they moved from robust to pre-frail, pre-frail to frail, stayed frail, and so on.
The findings show that older diabetic adults who deteriorated from pre-frail to frail had about twice the risk of developing CVD compared to those who remained robust throughout the study period. Similarly, those who remained persistently frail also faced about twice the risk. Even those who went from being robust to becoming pre-frail or frail had a 60% higher risk. Importantly, these elevated risks held up even after accounting for the possibility that some participants might die from non-CVD causes before experiencing a CVD event—a statistical challenge known as 'competing risks.'
This research suggests that monitoring changes in frailty status over time—not just measuring frailty at a single point—may be important for identifying older diabetic individuals at higher risk of cardiovascular disease. The consistency of findings across two culturally and geographically distinct cohorts (China and England) adds confidence to these associations. Early identification of those transitioning toward or remaining in frailty could be relevant for clinical monitoring and preventive strategies in elderly diabetic populations.
Li Y, Li S, Jin A. (2026). [A prospective dual-cohort study on the association between frailty index outcomes and cardiovascular disease onset risk in elderly diabetic populations].. Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine]. https://doi.org/10.3760/cma.j.cn112150-20251016-00994