DENVER — The FRAX 10-year fracture risk tool was fairly accurate in predicting the observed number of hip fractures that occurred among more than 5,000 participants of the Framingham Heart Study, according to data presented as a poster at the annual meeting of the American Society for Bone and Mineral Research.
The 10-year observed incidence of hip fracture for women was 117 cases, which did not differ significantly from the FRAX predicted number of 113. For men, the observed incidence was 29 cases, also not significantly different from the FRAX predicted number of 38, reported Elizabeth J. Samelson, Ph.D., of the Institute for Aging Research in Boston, and her coinvestigators.
FRAX is an online tool that was developed by the World Health Organization to calculate the 10-year probability of hip fracture and major osteoporotic fracture in women and men, aged 40–90 years, on the basis of bone mineral density (BMD), sex, age, smoking status, glucocorticoid use, height and weight, diagnosis of rheumatoid arthritis or secondary osteoporosis, history of fracture, and parental history of fracture. FRAX was developed using several population-based cohorts.
This study included 5,204 Framingham cohort members (2,917 women and 2,287 men) who had a baseline examination in 1987–2001 and were followed for hip fracture over 10 years. All were white. At baseline, patients were assessed for age, body mass index, current smoking status, alcohol consumption, glucocorticoid use, diagnosis of rheumatoid arthritis, prior fragility fracture, parental history of fracture, and T score. History of parental hip fracture was not available for members of the original cohort (1,456); these participants were classified as having no parental history of hip fracture. Femoral neck BMD was available for 4,224 participants.
The researchers used FRAX version 3.0 to calculate the 10-year probability of hip fracture and compared the expected number with the number observed in the cohort. A hip fracture was defined as a proximal femur fracture and was confirmed by review of medical records (including radiographic and surgical reports). Data were further analyzed by age and sex.
Among women aged 40–75 years, the incidence was 52 cases, compared with 57 expected by FRAX; among men aged 40–75 years, the incidence was 12 cases, compared with 23 expected by FRAX. Notably, the observed probability of hip fracture in the oldest adults (aged 76–90 years) exceeded the number predicted by FRAX, while the opposite was true for those aged 40–75. However, these differences were not significant, the authors noted. Among women aged 76–90 years, the incidence was 65 cases, compared with 55 expected by FRAX; among men aged 76–90 years, the incidence was 17 cases, compared with 14 expected by FRAX.
FRAX can be useful to communicate osteoporosis risk in white U.S. adults in the clinical or public health setting; “however, the tool may overestimate hip fracture risk in persons aged 40–75 years,” the researchers wrote.
The latest version of FRAX can be accessed at www.shef.ac.uk/FRAX
A related video is at www.youtube.com/InternalMedicineNews