Predicting fracture risk in osteoporosis: the use of fracture prediction tools in an osteoporosis clinic population

Background In the UK, the National Institute for Health and Care Excellence recommends either fracture risk assessment tool (FRAX) or QFracture to estimate the 10 year fracture risk of individuals. However, it is not known how these tools compare in determining risk and subsequent treatment using set intervention thresholds or guidelines. Methods The 10 year major osteoporotic (MO) and hip (HI) fracture risks were calculated for 100 women attending osteoporosis clinic in 2010 using FRAX and QFracture, and subsequent agreement to treatment between the tools was looked at using National Osteoporosis Foundation and National Bone Health Alliance thresholds (FRAX-20/3 and QFracture 20/3). We also looked at using these thresholds for QFracture and comparing them with the National Osteoporosis Guideline Group (NOGG) guidelines for FRAX (FRAX-NOGG). Results The 10 year risk for MO fracture for FRAX was 17.0% (IQR 10.8-24.0) and that of QFracture was 15.8% (IQR 9.5-27.7) (p=0.732). The 10 year risk for HI fracture for FRAX was 5.0% (IQR 2.1-8.9) and that of QFracture was 8.1% (IQR 2.5-21.6) (p<0.001). The agreement between FRAX-20/3 and QFracture-20/3 was greater than the agreement between FRAX-20/3 and FRAX-NOGG or QFracture-20/3 and FRAX-NOGG. Conclusions The calculated 10 year risk for MO fracture between FRAX and QFracture was similar, whereas that of HI fracture was significantly different. The agreement to treatment between QFracture-20/3 and FRAX-NOGG was only 45%. Treatment decisions can differ depending on the fracture calculation tool used when coupled with certain intervention thresholds or guidelines.