The fracture risk assessment tool (FRAX®) estimates the 10-year risk of hip fracture and major osteoporotic fracture in individual patients. Kanis et al. now report that a revised FRAX® model (version 3.0) gives reduced fracture probability estimates in patients 50 and/or 60 years old. “The revised FRAX® model ... does not alter the ranking of fracture probabilities but provides lower probability estimates ... in younger women and men,” write the authors.

FRAX® is an algorithm that computes the fracture probability in a patient by collating multiple clinical risk factors. Declining hip fracture incidence and decreasing age-adjusted mortality in the white US population led the National Osteoporosis Foundation to revise the FRAX® model by removing incidental or nonclinical vertebral fractures and incorporating updated US mortality data.

The researchers found a correlation between fracture probability estimates derived from the two FRAX® versions, although the revised version gave somewhat lower values. For example, in women aged 50 years and 60 years the median probability of hip fracture was 43% and 30% lower, respectively, when calculated by FRAX® 3.0 than when using FRAX® 2.0. In men of these ages, FRAX® 3.0 gave median values 40% and 27% lower than FRAX® 2.0. In all patients aged ≥70 years, both models gave similar fracture probabilities.

“These changes should improve the estimates of fracture probability,” say the investigators. However, they stress that the revisions to the FRAX® model do not change the rank order of fracture probability. “The revisions made have little impact on the efficiency with which the FRAX® tool categorizes risk ... because ... an individual at the 90th percentile of risk [calculated by FRAX® 2.0] would still be at the 90th percentile of risk using the revised FRAX® tool,” conclude the researchers.