Smoking behavior and frailty: joint impact on incident cardiovascular disease in a prospective cohort study

Why clinicians should know about it

• Picked for Epidemiology (top studies of the week, 8 November 2026): Large cohort examines smoking, frailty, interaction, and cardiovascular risk

Abstract

BACKGROUND: This study aimed to investigate the joint associations of smoking behaviors and frailty phenotype with incident cardiovascular disease (CVD), and to examine how associations of smoking cessation duration with CVD vary across different frailty states. METHODS: We included 297,340 UK Biobank participants free of baseline CVD, with baseline assessment of smoking behaviors (including smoking status and years since quitting for former smokers) and frailty phenotypes (i.e., nonfrail, prefrail, and frail). Cox proportional hazards models were used to evaluate the associations of combined frailty-smoking categories with coronary artery disease (CAD), stroke, and four secondary cardiovascular outcomes. Both additive and multiplicative interactions between frailty and smoking were evaluated using relative excess risk due to interaction (RERI), attributable proportion, synergy index, and interaction terms in Cox models. RESULTS: After a median follow-up period of 13.5 (IQR: 12.7-14.2) years, 20,621 participants (6.9%) developed incident CAD and 5,696 (1.9%) experienced stroke. Joint analyses demonstrated that current smokers with frailty consistently exhibited the highest cardiovascular risk across outcomes, with the strongest joint associations observed for peripheral artery disease (PAD) and heart failure (HF). Significant positive additive interactions between frailty and smoking were also identified for CAD (RERI = 0.20, 95% CI: 0.004-0.39) and PAD (RERI = 1.43, 95% CI: 0.70-2.20), indicating that the coexistence of smoking and frailty was associated with excess cardiovascular risk beyond the sum of their individual associations. Longer cessation duration was consistently associated with progressively lower cardiovascular risk across frailty categories. Significant multiplicative interactions between frailty and smoking were observed for CAD, PAD, and cardiovascular mortality. Nonfrail and prefrail individuals showed lower CAD and PAD risk within 5 years after smoking cessation, whereas frail individuals had persistently elevated risk during 5-15 years before approaching that of never smokers. No multiplicative interaction was observed for stroke, HF, or atrial fibrillation (AF), although longer smoking cessation duration remained associated with lower cardiovascular risk across all frailty strata. Finally, smoking-related cardiovascular risk varied across specific frailty components, particularly exhaustion, low physical activity, and slow gait speed. CONCLUSION: Frailty may increase susceptibility to smoking-related cardiovascular harm, particularly for CAD and PAD. Frail smokers may represent a particularly vulnerable population that warrants earlier smoking cessation support and more intensive cardiovascular prevention strategies to mitigate long-term cardiovascular risk. Joint consideration of smoking exposure and frailty status may therefore have greater public health relevance for reducing absolute cardiovascular risk burden than addressing either risk factor alone.

Abstract as published, via PubMed.