Individualized fracture risk assessment: progresses and challenges

Purpose of review Fragility fracture is a major public health burden, because it is associated with a substantial morbidity and mortality. Risk prediction models, including the Fracture Risk Assessment Tool (FRAX) and Garvan Fracture Risk Calculator (GFRC), have been developed to provide a useful clinical framework for communicating the risk of fracture. The present review examines the validation of risk prediction models in osteoporosis and identifies some major challenges. Recent findings Recent validation studies suggested that the area under the ROC curve in fracture discrimination ranged from 0.61 to 0.83 for FRAX, and from 0.63 to 0.88 for GFRC, with hip fracture having a better discrimination than fragility fractures as a group. FRAX substantially underestimated the risk of fracture, whereas the predicted risk by GFRC was close to or slightly higher than the actual risk. Results of post-hoc analyses of clinical trials indicated the antifracture efficacy of alendronate, coronate, bazedoxifene, and denosumab was greater in patients with higher predicted risk of fracture. However, there was no correlation between antifracture efficacy and predicted fracture risk among patients on raloxifene and strontium ranelate. Summary The prognostic performance of FRAX and GFRC for fracture prediction is not perfect, but these predictive models can aid patients and doctors to communicate about fracture risk in the medium term and to make rational decisions. However, the application of these predictive models in making decisions for an individual should take into account the individual's perception of the importance of the risk of fracture and its severity outcomes.