Updated fracture incidence rates for the Italian version of FRAX®

Abstract

In order to update data underlying the Italian version of FRAX, we computed the national hip fracture incidence in Italy from hospitalization records for the year 2008. Mortality data and 10-year probabilities of major osteoporotic fractures were also updated. This revision will improve FRAX accuracy and reliability.

Introduction: The original Italian version of FRAX® was based on five regional estimates of hip fracture risk undertaken up to 20 years previously. Our objective was to update hip fracture rates for the model with more recently derived data from the whole Italian population and more recent data on mortality.

Methods: We analyzed the Italian national hospitalization database for the year 2008 in order to compute age- and sex-specific hip fracture incidence rates. Re-hospitalisations of the same patients within 1 year were excluded from the analysis. Hip fracture incidence rates were computed for the age range of 40-100 years, whereas the original FRAX model lacked data on the youngest and oldest age groups. In addition, we used the national mortality data for the same year 2008 to update the model. Ten-year fracture probabilities were re-calculated on the basis of the new fracture incidence rates.

Results: The new hip fracture age- and sex-specific incidence rates were close to those used in the original FRAX tool, although some significant differences (not exceeding 25-30 %) were found for men aged 65-75 years and women under 55 years of age. In general, the revision resulted in decreased estimated 10-year probabilities in the younger age groups, whilst those in the older age groups were slightly increased.

Conclusions: The Italian version of FRAX has been updated using the new fracture incidence rates. The impact of these revisions on FRAX is likely to increase the accuracy and reliability of FRAX in estimating 10-year fracture probabilities.