History of fractures as predictor of subsequent hip and nonhip fractures among older mexican americans

To examine the association between previous fracture and risk of new hip and nonhip fractures over a seven-year period among older Mexican Americans. Data used are from the Hispanic Established Population for the Epidemiological Study of the Elderly (H-EPESE) (1993-2001). Measures included history of previous fracture (hip fracture only, a nonhip fracture, hip and nonhip fractures, and no fractures), sociodemographic factors, smoking status, medical conditions (arthritis, diabetes, stroke and cancer), activities of daily living disability, and high depressive symptoms. Cox proportional regression model was used to estimate the seven-year incidence of fractures. Of the 2,589 subjects, 42 reported a hip fracture, 328 reported a nonhip fracture, and 2,219 did not report a fracture at baseline. After controlling for all covariates, the hazard ratio (HR) of new hip fracture at seven-year follow-up was 6.48 (95% CI: 3.26-12.97) for subjects with only hip fracture at baseline and 1.96 (95% CI: 1.22-3.16) for subjects with nonhip fracture at baseline. The HR of new nonhip fracture was 1.90 (95% CI: 0.96-3.77) for subjects with only hip fracture at baseline and 2.62 (95% CI: 1.95-3.52) for subjects with nonhip fracture at baseline. A previous history of fractures in older Mexican Americans is the strongest predictor of recurrent fractures at hip and nonhip sites, independent of other health measures. Our findings of recurrent fractures suggest the need for more aggressive detection and adequate treatment of osteoporosis- and fall-related factors in this population.