Abstract / Summary
Abstract Background Insulin resistance is implicated in the development of cardiometabolic multimorbidity (CMM), but evidence regarding the associations of baseline, cumulative, and longitudinal patterns of estimated glucose disposal rate (eGDR) with incident CMM remains limited. We investigated these associations and evaluated the incremental predictive value of eGDR using two prospective cohorts. Methods Data were obtained from the English Longitudinal Study of Ageing (ELSA) and the Health and Retirement Study (HRS). Baseline eGDR analyses included 14,970 participants, while cumulative and trajectory analyses included 5,854 participants with repeated eGDR measurements. Group-based trajectory modeling was used to identify longitudinal eGDR trajectories. Cox proportional hazards and restricted cubic spline (RCS) models were applied to evaluate associations with incident CMM, defined as the coexistence of at least two of heart disease, stroke, and diabetes. Cohort-specific estimates were combined using random-effects meta-analysis. Predictive performance was assessed using time-dependent area under the receiver operating characteristic curve, C-index, net reclassification improvement, and integrated discrimination improvement. Subgroup and sensitivity analyses were also performed. Results During follow-up, 2,218 and 526 incident CMM events occurred in the baseline and cumulative eGDR cohorts, respectively. Each 1-SD increase in baseline eGDR was associated with a 40% lower risk of CMM (pooled HR 0.60, 95% CI 0.57–0.64). Compared with Q1, Q4 was associated with a pooled HR of 0.27 (95% CI 0.22–0.32; P for trend < 0.001). For cumulative eGDR, HRs per 1-SD increase were 0.68 (95% CI 0.54–0.84) in ELSA and 0.52 (95% CI 0.45–0.60) in HRS; corresponding HRs for Q4 versus Q1 were 0.44 (95% CI 0.23–0.85) and 0.15 (95% CI 0.09–0.24), respectively (P for trend < 0.05). Four eGDR trajectories were identified. Compared with the persistently low trajectory, the persistently high trajectory was associated with a 79% lower risk of CMM (pooled HR 0.21, 95% CI 0.10–0.41; P for trend < 0.01). RCS analyses indicated inverse and linear associations of baseline and cumulative eGDR with CMM risk. Adding baseline eGDR modestly improved longer-term discrimination and improved reclassification over the base model, while gains from cumulative eGDR and its trajectories were more consistent in HRS. eGDR did not outperform models incorporating its individual components separately. Subgroup and sensitivity analyses generally corroborated the primary findings. Conclusions Higher baseline and cumulative eGDR and a persistently high eGDR trajectory were associated with substantially lower CMM risk in middle-aged and older adults. These findings support the potential value of incorporating both single and repeated eGDR measurements into CMM risk assessment.