Social Determinants of Health (SDoHs) exert their influence primarily through social structures and environmental conditions. However, limited evidence exists regarding which specific SDoHs are associated with all-cause and cardiovascular mortality, and whether incorporating SDoHs into the Life's Essential 8 (LE8) framework improves mortality prediction. Data were analyzed from the 2005-2018 National Health and Nutrition Examination Survey (NHANES). Weighted Cox proportional hazards models were used to estimate associations between SDoHs and mortality. Additive interaction effects were quantified using the relative excess risk due to interaction (RERI). Predictive performance was assessed using continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) for all-cause and cardiovascular mortality. Among 22,938 participants, 2,344 all-cause deaths (7.29%) and 720 cardiovascular deaths (2.12%) were observed over a median follow-up of 7.83 years. Multivariable analyses revealed that unemployment (HR = 1.44, 95% CI: 1.25-1.66), low-income status (poverty-to-income ratio (PIR) < 1; HR = 1.75, 95% CI: 1.32-2.31), severe food insecurity (HR = 1.45, 95% CI: 1.19-1.78), and being unmarried or living alone (HR = 1.43, 95% CI: 1.28-1.59) were independent predictors of all-cause mortality (all P < 0.001). Low PIR (HR = 1.90, 95% CI: 1.21-2.97) and being unmarried or living alone (HR = 1.52, 95% CI: 1.26-1.85) were also significantly associated with cardiovascular mortality. Significant additive interactions were found between unemployment and low PIR (RERI = 0.90), food insecurity (RERI = 0.59), and unmarried or living alone (RERI = 0.37) for all-cause mortality. An additive interaction between low PIR and unmarried or living alone (RERI = 0.46) was observed for cardiovascular mortality. Including these SDoHs significantly improved 1-, 3-, 5-, and 10-year prediction of all-cause and cardiovascular mortality (NRI and IDI, all P < 0.05). Unemployment, low PIR, severe food insecurity, and being unmarried or living alone were identified as independent predictors of increased all-cause mortality. Low PIR and being unmarried or living alone were also associated with cardiovascular mortality. Incorporating these SDoHs into the LE8 framework improved predictive performance for both mortality outcomes.
Zhuang et al. (Mon,) studied this question.