Given the recent availability of FRAX (www.shef.ac.uk/FRAX) and the National Osteoporosis Guideline (www.shef.ac.uk/NOGG), we were pleased to see the paper by Hippisley-Cox and Coupland that draws attention to the need for better targeting of treatment to patients at risk of fracture. We would contend, however, that the rather complex QFracture score adds little, if any, value to the present position. Despite limitations in the comparison with the much simpler FRAX tool (e.g. comparing an estimated incidence from Kaplan-Meier analysis with a probability), the study provided an excellent validation of the FRAX tool particularly bearing in mind that the latter has been derived and validated internationally.
The authors also acknowledge that the Qfracture score also lacks input of two of the major risk factors for fracture, especially hip fracture, namely a measurement of femoral neck bone mineral density (BMD) and prior fracture. To dismiss BMD measurements as “expensive and inconvenient tests” dismisses a wealth of data demonstrating a much higher gradient of risk for BMD (e.g. a 2.6 fold increase in fracture risk per 1 SD decrease in femoral neck BMD that is also age-dependent) than the risk factors included in the QFracture questionnaire.
Furthermore, the absence of BMD measurements in younger patients is not compliant with NICE and NOGG guidance and would cause significant problems in PCT audit. Prior fracture alone is the leading single risk factor of utility in any case-finding strategy for the management of osteoporosis as outlined in the NOGG guideline – it’s absence from QFracture severely limits its clinical value. As stated by the authors, assessment tools should identify those individuals at high risk who might best benefit from an intervention. We are not aware of studies that demonstrate that high risk individuals identified by many of the Qfracture risk factors (cardiovascular disease, type 2 diabetes, asthma, tricyclic antidepressants usage, history of falls or liver disease) characterises a risk that is amenable to bone targeted interventions. In contrast, patients at high risk identified by FRAX have been shown to respond to pharmacological interventions.
The FRAX tool and in particular the NOGG guideline for its clinical implementation (available by a direct link from the FRAX UK calculation website) have been widely endorsed by professional and lay organisations within the NHS including the Royal College of Physicians and the Primary Care Rheumatology Society. In contrast to the assertion that “FRAX is not currently in widespread use in primary care in the United Kingdom”, the NOGG website has received over 1 million hits since its launch one year ago and receives an average of 4500 hits daily. We readily recognise that it will be more widely used when incorporated into GP software systems and hope that this will become a reality in the very near future.
Competing interests:
Developers of FRAX and contributors to NOGG
Rapid Response:
Limited application of QFracture
Given the recent availability of FRAX (www.shef.ac.uk/FRAX) and the National Osteoporosis Guideline (www.shef.ac.uk/NOGG), we were pleased to see the paper by Hippisley-Cox and Coupland that draws attention to the need for better targeting of treatment to patients at risk of fracture. We would contend, however, that the rather complex QFracture score adds little, if any, value to the present position. Despite limitations in the comparison with the much simpler FRAX tool (e.g. comparing an estimated incidence from Kaplan-Meier analysis with a probability), the study provided an excellent validation of the FRAX tool particularly bearing in mind that the latter has been derived and validated internationally.
The authors also acknowledge that the Qfracture score also lacks input of two of the major risk factors for fracture, especially hip fracture, namely a measurement of femoral neck bone mineral density (BMD) and prior fracture. To dismiss BMD measurements as “expensive and inconvenient tests” dismisses a wealth of data demonstrating a much higher gradient of risk for BMD (e.g. a 2.6 fold increase in fracture risk per 1 SD decrease in femoral neck BMD that is also age-dependent) than the risk factors included in the QFracture questionnaire.
Furthermore, the absence of BMD measurements in younger patients is not compliant with NICE and NOGG guidance and would cause significant problems in PCT audit. Prior fracture alone is the leading single risk factor of utility in any case-finding strategy for the management of osteoporosis as outlined in the NOGG guideline – it’s absence from QFracture severely limits its clinical value. As stated by the authors, assessment tools should identify those individuals at high risk who might best benefit from an intervention. We are not aware of studies that demonstrate that high risk individuals identified by many of the Qfracture risk factors (cardiovascular disease, type 2 diabetes, asthma, tricyclic antidepressants usage, history of falls or liver disease) characterises a risk that is amenable to bone targeted interventions. In contrast, patients at high risk identified by FRAX have been shown to respond to pharmacological interventions.
The FRAX tool and in particular the NOGG guideline for its clinical implementation (available by a direct link from the FRAX UK calculation website) have been widely endorsed by professional and lay organisations within the NHS including the Royal College of Physicians and the Primary Care Rheumatology Society. In contrast to the assertion that “FRAX is not currently in widespread use in primary care in the United Kingdom”, the NOGG website has received over 1 million hits since its launch one year ago and receives an average of 4500 hits daily. We readily recognise that it will be more widely used when incorporated into GP software systems and hope that this will become a reality in the very near future.
Competing interests:
Developers of FRAX and contributors to NOGG
Competing interests: No competing interests