U-shaped associations between cardiometabolic index (CMI) and all-cause and cardiovascular mortality among elderly Americans with diabetes or prediabetes: NHANES 1999–2018

Background: As a relatively novel and integrated metric, the Cardiometabolic Index (CMI) has not been thoroughly explored for its utility in assessing mortality risk among patients with diabetes or prediabetes. This study focuses on analyzing the relationship between CMI and all-cause along with cardiovascular disease (CVD) mortality in older U.S. adults diagnosed with diabetes or prediabetes. Methods: The research encompassed 3,062 older adults diagnosed with diabetes or prediabetes, drawn from the National Health and Nutrition Examination Survey (NHANES) conducted between 1999 - 2018. The mortality results were ascertained by cross-referencing with the records of the National Death Index (NDI) up until December 31, 2019. Moreover, the DeLong test was utilized to validate the predictive capability of CMI relative to other cardiometabolic indices. Results: Across a follow-up period averaging 85.44 months, 1,277 all-cause deaths were documented among patients with diabetes or prediabetes, including 445 occurrences of cardiovascular mortality. The analysis of threshold effects based on restricted cubic splines demonstrated a U-shaped, nonlinear association between CMI and all-cause as well as CVD mortality in individuals with diabetes or prediabetes. Notably, Interestingly, when the baseline CMI was below the cutoff values (1.12 for all-cause mortality and 1.04 for cardiovascular mortality), it showed a negative association solely with all-cause mortality (HR: 0.78, 95% CI: 0.63–0.95). However, exceeding these thresholds was significantly linked to a higher risk of both all-cause mortality (HR: 1.13, 95% CI: 1.03–1.24) and CVD mortality (HR: 1.19, 95% CI: 1.03–1.37). Conclusion: In older adults with diabetes or prediabetes, a U-shaped association was identified between initial CMI and both all-cause and CVD mortality, with respective thresholds of 1.12 and 1.04.