Body mass index and a body shape index: Individual and interactive predictors of all-cause and cardiovascular mortality risk in depression

OBJECTIVE: Conflicting evidence exists on the association of obesity, depression and mortality risk. This study evaluates a cohort of subjects with depression from the USA NHANES and all-cause and cardiovascular mortality risks based on baseline BMI and ABSI. METHODS: A nationally representative cohort study was conducted using NHANES data (2005-2018), including 2854 adults with depression. Weighted Cox regression and restricted cubic spline (RCS) models were applied to assess anthropometric associations with mortality. Weighted receiver operating characteristic (ROC) curves evaluated the diagnostic value of BMI and ABSI. RESULTS: Each 1-SD BMI increase was predictive of reduced all-cause (HR = 0.81, 95 % CI:0.67-0.98) and cardiovascular mortality (HR = 0.62, 95 % CI:0.43-0.89). Each 1-SD ABSI increase was predictive of elevated all-cause (HR = 1.27, 95 % CI:1.06-1.51) and cardiovascular mortality (HR = 1.60, 95 % CI:1.20-2.15). In these depressed individuals, BMI showed the U-shaped and ABSI the expected curvi-linear relationships with mortality as established from the studies of the NHANES cohorts. In ROC analyses, ABSI showed the best predictive performance for all-cause mortality (AUC = 0.678) and for cardiovascular mortality (AUC = 0.706). BMI Tertile 1 & ABSI Tertile 3 had the highest mortality risks (all-cause HR = 3.09, 95 % CI:1.38-6.93; cardiovascular HR = 5.45, 95 % CI:1.36-21.85). Due to the approximate statistical independence of BMI and ABSI, combining them further improved AUC values (all-cause: 0.691; cardiovascular: 0.722). CONCLUSION: Among individuals with depression ABSI predicts mortality similar to the general population, while obesity as defined by BMI was "paradoxically" associated with lower mortality. Personalized mortality risk for an individual can be derived by combining the independent risks based on BMI and ABSI.