Fracture Risk Models

The key to addressing the impact of osteoporosis and is to identify patients at risk of the disease prior to fracture and to ensure that these patients are appropriately screened and treated if necessary. DXA can be used to diagnose osteoporosis^1-3. However, it can also be used to predict fracture risk^4-10, since the risk of fracture increases as BMD decreases in men and women⁶. Low BMD is the single best predictor of future fracture in postmenopausal women without prior fracture^3,10 and has similar predictive potential in elderly men⁹. Fractures are multifactorial in origin^2-4,7-10, however, so identifying persons at greater risk involve direct patient assessment using knowledge of risk factors for fractures and falls. Persons with multiple risk factors for fracture are at much higher risk of future fracture than persons with any single risk factor, including low BMD^7-10. Combining BMD with other risk factors such as age or prior fracture, or combining multiple risk factors greatly enhances the ability to predict persons or populations at increased risk for future fractures^7-10. This is important since many persons who fracture do not have osteoporosis by DXA criteria^10,11. Independent predictors of future fracture identified from multiple studies include prior fracture, low BMD, fall, advancing age, glucocorticoid use and frailty^4,6-12. Fracture risk tools have thus been developed which may assist clinicians to identify groups of patients, or individuals, at increased risk of fracture^13-15. By considering multiple risk factors when assessing fracture risk, rather than just BMD, the number of persons needing treatment to prevent a single fracture could be dramatically reduced from several thousand to less than one hundred¹⁶.

Risk factors have limitations for assessing risk and prognosis^17-19 highlighting the difficulties involved in trying to achieve effective risk stratification in multifactorial disease processes such as osteoporosis^4,12. Although fracture risk models predict populations at increased risk of fracture, the result for an individual person is usually an average number around which there is usually considerable variation^7,12. Although fracture risk models are interesting tools, each model has shortcomings which need to be taken into account when evaluating individual patients with their results.

Some of the short-comings of FRAX^® have been identified and corrected; for instance: elimination of the Z-score from the tool and more recently substitution of the T-score with the manufacturers actual BMD. Similar to diagnostic tests, the internal and external validity, sensitivity and specificity of these models need to be considered. Although fracture risk models may have high internal validity²⁰, recently presented results of a study evaluating the external validity of FRAX using data from the Women’s Health Initiative were disappointing showing the c-statistic to be around 0.621. While studies differ, they do suggest that risk factors for fracture differ between women of different ethnicities⁸, and between men and women^7-9, adding complexity to the assessment of fracture risk. This has resulted in the development of different fracture risk assessment tools^13-15. To facilitate ease of use, presently available risk models are parsimonious and thus no tool will incorporate all possible risk factors. Other issues concerning FRAX have already been highlighted with specific examples²², and further discussion will take place at the next ISCD Position Development Conference in 2010²³.

The information generated with the use of these tools may assist physicians in making decisions about who to treat, especially for patients with osteopenia. However, there are no published studies showing that any fracture model clearly supersedes a clinician’s own judgment. In addition, prior studies have shown that some elderly women have a very high 1-year risk of fracture – around 25% – thus a prediction model is not needed in such instances²⁴. Thus fracture prediction models should be used judiciously in managing individual patients.

Suggested References:

1. ISCD Official Positions. Available at: http://staging.iscd.org/official-positions
2. Report of a WHO Study Group. 1994 Assessment of Fracture Risk and its Application to Screening for Postmenopausal Osteoporosis. WHO Technical Report Series 843.
3. NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis and Therapy. 2001 Osteoporosis prevention, diagnosis and therapy. JAMA 285(6):785-795.
4. Report of a WHO Scientific Group. 2003 Prevention and Management of Osteoporosis. WHO Technical Report Series 921.
5. Introduction to Understanding Bone Densitometry.
6. Marshall D, Johnell O, Wedel H. 1996 Meta-analysis of how well measures of bone mineral density predict occurrences of osteoporotic fractures. BMJ 312(7041):1254-1259.
7. Cummings SR, Nevitt MC, Browner WS, et al. 1995 Risk factors for hip fracture in white women. Study of osteoporotic fractures research group. N Engl J Med 332(12):767-773.
8. Cauley JA, Wu L, Wampler NS, et al. 2007 Clinical risk factors for fractures in multi-ethnic women: the Women’s Health Initiative. J Bone Miner Res 22(11):1816-1826.
9. Lewis CE, Ewing SK, Taylor BC, et al. 2007 Predictors of non-spine fractures in elderly men: the MrOS study. J Bone Miner Res 22(2): 211-219.
10. Siris ES, Chen YT, Abbott TA, et al. 2004 Bone mineral density thresholds for pharmacological intervention to prevent fractures. Arch Intern Med 164(10):1108-1112.
11. Wainwright SA, Marshall LM, Ensrud KE, et al. 2005 Hip fracture in women without osteoporosis. J Clin Endocrinol Metab 90(5): 2787-2793.
12. Kanis JA, Burlet N, Cooper C, et al. 2008 European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 19(4): 399-428.
13. Black DM, Steinbuch M, Palermo L, et al. 2001 An assessment tool for predicting fracture risk in postmenopausal women. Osteoporos Int 12(7):519-528.
14. FRAX: Fracture Risk Assessment tool. Available at: http://www.shef.ac.uk/FRAX/ Accessed 30th March 2009.
15. Med Surf: absolute risk and remaining lifetime fracture probability. Accessed 30th March 2009.
16. Nelson, HD, Helfand M, Woolf SH, Allan JD. 2002 Screening for postmenopausal osteoporosis: a review of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 137(6):529-541.
17. Ware JH. 2006 The limitations of risk factors as prognostic tools. N Engl J Med 355(25):2615-2617.
18. Fletcher RH, Fletcher SW, Wagner EH. 1996 Clinical Epidemiology: The Essentials. Williams & Wilkins, Baltimore, MD.
19. Gordis L. 2004 Epidemiology. Elsevier Saunders, Philadelphia, PA.
20. Hans D, Durosier C, Kanis JA, et al. 2008 Assessment of the ten-year probability of osteoporotic hip fracture combining clinical risk factors and heel bone ultrasound: the EPISEM prospective cohort of 12,958 elderly women. J Bone Miner Res 23(7):1045-1051.
21. Cauley JA. Worldwide Epidemiology of Fractures: Difference Between East and West. 2009 Proceedings of ISCD Annual Meeting, Orlando, Fla, USA.
22. Watts NB, Lewiecki EM, Miller PD, Baim S. 2008 National Osteoporosis Foundation 2008 Clinician’s Guide to Prevention and Treatment of Osteoporosis and the World Health Organization Fracture Risk Assessment Tool (FRAX): what they mean to the bone densitometrist and bone technologist. J Clin Densitom 11(4):473-7.
23. ISCD Position Development Conference Bucharest 2010.
24. Lindsay R, Silverman SL, Cooper C, et al. 2001 Risk of new vertebral fracture in the year following a fracture. JAMA 285(3):320-323.