Status
Completed
Title
Derivation and validation of the QFracture algorithm to predict risk of osteoporotic fracture
What were the objectives of the study?
QFracture is a risk prediction algorithm which calculates an individual’s risk of osteoporotic fracture taking account of their individual risk factors such as age, sex, ethnicity, clinical values and diagnoses. Osteoporotic fracture is identified according to clinical codes recorded on the primary care record or the linked mortality or hospital record. The research describing the derivation and validation of different versions of QFracture has been published in the BMJ.
QFracture is used to quantify an individual’s risk of having an osteoporotic fracture in the next 10 years in order to prioritise and inform decisions regarding interventions to lower risk. It is recommended by NICE as the risk score for use in its guidance on prevention of fragility fracture. It is used as a measure in the GP contract. It is updated regularly to ensure the algorithms remain calibrated to the current population since the incidence of disease and the prevalence of risk factors change over time. The updates also enable the algorithm to take account of improvements to data quality (for example increases in the completeness of recording of risk factors such as ethnicity or laboratory values) and also allows adaptations to any changes which might be required to improve the predictive power of the tool or the population to which is applied.
How was the research done?
Cox proportional hazards models and cause specific hazard models in the derivation cohort are used to derive separate risk equations in men and women. These equations predict both 1-10 year risk and lifetime risk. The equations are regularly updated to take account of changes in population risks over time, availability of new data, improvement in statistical methods and changes in requirements for how the tools can be used in clinical practice.
Measures of calibration, discrimination and reclassification are determined in the validation cohort for men and women separately and for individual subgroups e.g. age group, ethnicity. External validation is conducted in separate databases such as CPRD and THIN by the authors and also by external researchers.
Chief Investigator
Julia Hippisley-Cox
Lead Applicant Organisation Name
Sponsor
Oxford
Location of research
Oxford
Date on which research approved
01-Feb-2019
Project reference ID
Q110/OX110
Generic ethics approval reference
18/EM/0400
Are all data accessed are in anonymised form?
Yes
Brief summary of the dataset to be released (including any sensitive data)
Risk factors for osteoporotic fracture linked to outcomes (hip, knee, wrist and spine fracture) on either GP, hospital or mortality records
What were the main findings?
QFracture works out the risk of a patient developing any osteoporotic fracture (i.e. hip, wrist, shoulder or spine) or hip fracture alone within the next 10 years taking account of their risk factors. It does not give a diagnosis but a risk. The calculation is based on variables stored in the patient’s electronic health record including age and sex as well as clinically relevant variables such as diagnoses, presenting symptoms and laboratory measurements. It works in two models.
1. It can be used by a clinician with an individual patient in the surgery setting to calculate their risk
2. It can be applied to the entire population in batch mode to generate an estimated risk for each patient so that the patients at highest risk can be prioritised for interventions and investigations
Implications and Impact
QFracture has been validated on separate populations of primary care patients and shown to have good performance at identifying those at high risk.
QFracture is recommended by NICE guidance and NICE Quality standards, Health Improvement Scotland and has been included in QOF. In 2017, a validation of QFracture reported improved performance compared with FRAX confirming the original work34 and excellent discrimination on three separate validations in CPRD.
In three studies (2015, 2017 and 2019), QFracture had the best discrimination for predicting hip fracture compared with FRAX and Garvan. QFracture has also been validated in subgroups (e.g. among people with Parkinson’s disease). QFracture has been validated internationally, for example, for use in Spain.
Funding Source
No external funding
Public Benefit Statement
Research Team
Julia Hippisley-Cox, Carol Coupland
Publications
-
Derivation and validation of updated QFracture algorithm to predict risk of osteoporotic fracture in primary care in the United Kingdom: prospective open cohort study
Authors: Hippisley-Cox J, Coupland C
Ref: BMJ 2012;344:e3427
https://www.bmj.com/content/344/bmj.e3427 -
Predicting risk of osteoporotic fracture in men and women in England and Wales: prospective derivation and validation of QFracture Scores
Authors: Hippisley-Cox J, Coupland C
Ref: BMJ 2009:339:b4229
https://www.bmj.com/content/339/bmj.b4229 -
QFracture 2016 Annual Update Information
Authors: Professor Julia Hippisley-Cox, Professor Carol Coupland
Ref:
https://www.qresearch.org/media/qirhz3ti/qfracture-2016-annual-update-information.pdf
Press Releases
- QFracture validated as a key tool for clinicians (12 July 2011)
Access Type
Trusted Research Environment (TRE)