Association between GDF15, poverty and mortality in urban middle-aged African American and white adults.

Mortality disparities are influenced by race and poverty. There is limited information about whether poverty influences biologic markers of mortality risk. Emerging data suggests that growth differentiation factor 15 (GDF15) is associated with mortality; however, the interplay between GDF15, sociodemographic factors and mortality is not known. We sought to evaluate the interactions between GDF15 and sex, race and poverty status on mortality. Serum GDF15 was measured in 1036 African American and white middle-aged men and women above and below 125% of the Federal poverty status from the Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) study. Multivariable adjusted Cox regression models were used to assess the association between log-transformed GDF15 (logGDF15) and 12-year mortality outcomes (all-cause, cardiovascular- and cancer-specific outcomes) and interactions with sex, race and poverty status. Likelihood ratio tests were used to assess significance of the interaction terms. Median GDF15 was 655.2 pg/mL (IQR = 575.1). During 12.2 years of follow-up, 331 died of which 94 cardiovascular- and 87 were cancer-specific deaths. One unit of increase in logGDF15 was associated with a hazard ratio for all-cause mortality, cardiovascular- and cancer-specific mortality of 2.26 (95% confidence interval [CI], 1.94-2.64), 2.74 (95%CI, 2.06-3.63) and 1.41 (95%CI, 1.00-2.00), respectively. There was an interaction between logGDF15 and poverty status on all-cause mortality (p<0.05). The GDF15×poverty status interaction term improved model calibration for all-cause mortality. Our study provides the first evidence that the effect of elevated GDF15 on all-cause mortality is modified by poverty status.