Abstract 116: BMI is Not a Risk Factor for All-Cause Mortality and Cardiovascular Disease Among African American Women

Abstract

Background: Across the age spectrum, obesity rates are higher among African Americans (AA) than Whites, particularly women. Previous findings from a large cohort study indicated that risk of all-cause mortality is only moderately elevated with increased weight among AA, but this relationship is a much stronger finding in Whites. In this study, we examined the relationship of baseline body mass index (BMI) to incident all-cause mortality and fatal and non-fatal cardiovascular and renal disease (CVRD) events. Methods: Between 1994 - 2006, 1723 AA (52% women, age: 44±7 yrs., BMI: 29 ± 6 kg/m 2 , 51% hypertensive) were recruited for genetic studies of hypertension. Height and weight were obtained at baseline in a standardized manner. Mortality data were obtained from the National Death Index in 2018 on all subjects. Incidence of fatal/non-fatal CVRD events was obtained and confirmed by review of paper and electronic medical records between 2015-2019 in a sub-sample. Subjects who had either a CVRD event or at least 10 years of follow-up data available with no event were included (n=788). Logistic regression (LR) was performed to determine if BMI predicted mortality or a CVRD event. Results: There were 325 deaths (19%, time-to-death: 12±3 yrs.) from all causes. During a follow-up of 14±4 yrs., 154 subjects had a CVRD (20%) event/death. There was no difference in age between sexes, but women had a higher BMI compared to men (30±7 vs. 27±5 kg/m 2 , p<0.0001). Men had a higher mortality (23% vs. 15%, p<0.0001) and more CVRD events (24% vs. 16%, p = 0.008). The odds ratio (OR) for death was 1.62 (CI: 1.3-2.1, p=0.0002) in men compared to women. Baseline BMI was not predictive of all-cause mortality in the total sample or in men or women separately. BMI predicted CVRD events in men with an OR of 1.08 (CI: 1.03-1.13), i.e. one-unit increase in BMI in men led to 8% increase in odds for CVRD events. However, BMI was not predictive of CVDR events in women (OR: 1.01 [0.98-1.05]). Conclusion: This is the first study to evaluate the effect of BMI longitudinally on incident CVRD events among AA. Although BMI is the most common metric to define obesity, it is not predictive of CVRD and/or mortality risk among AA women. This factor should be considered while determining cardiovascular disease risk status among AA.