Separate and combined value of bone mass and gait speed measurements in screening for hip fracture risk: results from the EPIDOS study. Epidémiologie de l'Ostéoporose

Osteoporos Int. 1999;9(2):188-92. doi: 10.1007/s001980050134.


Based on data from the EPIDOS prospective study, we have shown that femoral bone mineral density (BMD), calcaneal ultrasound measurements and fall-related factors are significant predictors of the risk of hip fracture. The goal of the present investigation, in the same cohort of elderly women, was (1) to assess and compare the value of femoral BMD, calcaneal broadband ultrasound attenuation (BUA), gait speed and age for identifying elderly women at high risk of hip fracture and (2) to determine whether combining two or more of these measurements would improve predictive ability over single measures. A total of 5895 elderly women had baseline measurements of femoral neck BMD by dual-energy X-ray absorptiometry, calcaneal BUA and gait speed. During an average of 33 months of follow-up, 170 women suffered a hip fracture. We compared the sensitivity and specificity of single and combined measures for three specific cutoff levels to define high risk, i.e., the median, the top quartile and the top decile of risk. We found that femoral BMD, calcaneal BUA, gait speed and age have approximatively the same discriminant value to identify women at high risk of hip fracture even though certain measures and combinations of measures have a significantly higher sensitivity for certain cutoff levels. The sensitivity of the available screening tools is low, even when they are combined: to obtain a sensitivity of about 80%, approximately 50% of the population must be considered to be at high risk.

Publication types

  • Comparative Study

MeSH terms

  • Aged
  • Bone Density*
  • Cohort Studies
  • Female
  • Femur Neck / physiology*
  • Gait / physiology*
  • Hip Fractures / prevention & control*
  • Humans
  • Osteoporosis, Postmenopausal / complications
  • Predictive Value of Tests
  • Prospective Studies
  • Sensitivity and Specificity
  • Time Factors