Associations of cumulative exposure and dynamic trajectories of the combined atherogenic-frailty index with incident cardiometabolic multimorbidity: a prospective cohort study

Background: Cardiometabolic multimorbidity (CMM) poses a growing global health burden, yet effective tools for early risk stratification remain limited. The combined impact of atherogenic dyslipidaemia and physiological frailty, captured by the atherogenic index of plasma-frailty index (AIP-FI), on CMM development is poorly understood. This study aimed to investigate the associations of cumulative exposure and longitudinal trajectories of AIP-FI with incident CMM. Methods: This prospective cohort study included 4,157 participants aged ≥45 years without baseline CMM from CHARLS. The AIP-FI was constructed as the product of AIP and a 37-deficit frailty index. Cumulative AIP-FI was calculated from repeated measurements. Longitudinal trajectories were identified using K-means clustering. Associations with incident CMM were evaluated by multivariable Cox models, RCS, K-M curves, ROC/NRI/IDI, mediation, interaction, subgroup, and sensitivity analyses. Results: Over a mean 8.5-year follow-up, 612 incident CMM cases occurred. Both higher baseline and cumulative AIP-FI were significantly associated with increased CMM risk, with cumulative exposure showing a stronger effect (fully-adjusted HR per SD increment: 1.46; 95% CI = 1.38-1.54). Participants in the highest quartile of cumulative AIP-FI had a near 4-fold higher CMM risk compared to the lowest (HR = 3.95; 95% CI = 2.95-5.29). Clustering analyses revealed that groups with high-stable AIP-FI exhibited the highest CMM risks. Cumulative AIP-FI demonstrated superior predictive performance for CMM (AUC = 71.17%) and significantly improved risk reclassification over traditional markers (NRI = 0.157). Mediation analyses suggested bidirectional synergistic effects between AIP and FI. Conclusions: In this national cohort, both the cumulative burden and adverse longitudinal trajectories of AIP-FI were independently associated with an elevated risk of CMM. Dynamic monitoring of this composite biomarker could facilitate the early identification of high-risk individuals, thereby informing targeted prevention strategies for middle-aged and older adults in China.