Abstract
OBJECTIVES: the objective of this study was to explore whether race-based difference in fall risk may be mediated by environmental and physical performance risk factors. METHODS: using data from a nationally representative longitudinal survey of 7,609 community-dwelling participants in the National Health and Aging Trends Study (NHATS), we evaluated whether racial differences in fall risk may be explained by physical performance level (measured by the Short Physical Performance Battery), mobility disability, physical activity level and likelihood of living alone. Multivariate Poisson regression and mediation models were used in analyses. RESULTS: in whites and blacks, the annual incidence of 'any fall' was 33.8 and 27.1%, respectively, and the annual incidence of 'recurrent falls' was 15.5 and 12.3%, respectively. Compared with whites, blacks had relative risks of 0.7 (95% confidence interval 0.6-0.8) and 0.6 (0.5-0.8) for sustaining any fall and recurrent falls, respectively, in adjusted analyses. Blacks had poorer performance on the SPPB (P < 0.001), higher levels of mobility disability (P < 0.001), similar levels of physical activity (P = 0.19) and were equally likely to live alone relative to whites (P = 0.77). Mediation analysis revealed that these risk factors collectively acted as suppressors and none of these factors accounted for the racial differences in fall risk observed. CONCLUSIONS: relative to whites, blacks were at 30 and 40% decreased risk of sustaining any fall and recurrent falls, respectively. This difference in risk remains unexplained.