Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

OBJECTIVE: To evaluate the added predictive accuracy of bone mineral density (BMD) to fracture risk assessment. DESIGN: Prospective cohort study using data between 01 January 2010 and 31 December 2012. SETTING: North Denmark Osteoporosis Clinic of referred patients presenting with at least one fracture risk factor to the referring doctor. PARTICIPANTS: Patients aged 40-90 years; had BMD T-score recorded at the hip and not taking osteoporotic preventing drugs for more than 1 year prior to baseline. MAIN OUTCOME MEASURES: Incident diagnoses of osteoporotic fractures (hip, spine, forearm, humerus and pelvis) were identified using the National Patient Registry of Denmark during 01 January 2012-01 January 2014. Cox regression was used to develop a fracture model based on predictors in the Fracture Risk Assessment Tool (FRAX®), with and without, binary and continuous BMD. Change in Harrell's C-Index and Reclassification tables were used to describe the added statistical value of BMD. RESULTS: Adjusting for predictors included in FRAX®, patients with osteoporosis (T-score ≤-2.5) had 75% higher hazard of a fracture compared with patients with higher BMD (HR: 1.75 (95% CI 1.28 to 2.38)). Forty per cent lower hazard was found per unit increase in continuous BMD T-score (HR: 0.60 (95% CI 0.52 to 0.69)).Accuracy improved marginally, and Harrell's C-Index increased by 1.2% when adding continuous BMD (0.76 to 0.77). Reclassification tables showed continuous BMD shifted 529 patients into different risk categories; 292 of these were reclassified correctly (57%; 95% CI 55% to 64%). Adding binary BMD however no improvement: Harrell's C-Index decreased by 0.6%. CONCLUSIONS: Continuous BMD marginally improves fracture risk assessment. Importantly, this was only found when using continuous BMD measurement for osteoporosis. It is suggested that future focus should be on evaluation of this risk factor using routinely collected data and on the development of more clinically relevant methodology to assess the added value of a new risk factor.

Original publication




Journal article


Bmj open

Publication Date





bone mineral density, fracture, osteoporosis, predictive accuracy, risk prediction, Adult, Aged, Aged, 80 and over, Bone Density, Denmark, Female, Humans, Male, Middle Aged, Osteoporotic Fractures, Predictive Value of Tests, Prospective Studies, Risk Assessment, Risk Factors