Gait speed and one-leg standing time each add to the predictive ability of FRAX

SUMMARY: Gait speed or one-leg standing time (OLST) as additional predictors in FRAX. Population 351 elderly women followed 10 years. Both could improve predictions. The area under curve (AUC) for FRAX is 0.59, OLST is 0.69 and gait speed is 0.71. The net reclassification index (NRI) for classification to highest risk quartile or lowest three quartiles was 0.24 for gait speed and non-significant for OLST. INTRODUCTION: The risk of falls and bone strength are two main determinants of hip fracture risk. The fracture risk assessment tool FRAX, however, lacks direct measures of fall risk(1). A short OLST and a slow gait speed are both fall-related risk factors for hip fractures. The aim of this study was to investigate whether the addition to FRAX of either gait speed or OLST could improve the predictive ability for hip fractures, compared to FRAX alone. METHODS: A population-based sample of 351 women aged between 69 and 79 years were tested for one-leg standing time with eyes open and mean gait speed over a 15 + 15-m walk. Fracture and mortality data were obtained from health care registers. RESULTS: The AUC for the receiver operating characteristic (ROC) increased from 0.61 to 0.71 when gait speed was added to FRAX. The AUC was 0.69 for OLST added to FRAX. The highest quartile of hip fracture risks according to FRAX had an absolute 10-year risk of ≥15%. The population was divided into one group with a hip fracture risk of ≥15% and one group with a fracture risk of <15%. NRI for addition of gait speed to FRAX was 0.24 (p = 0.023), while NRI was 0.08 (p = 0.544) for addition of OLST to FRAX. CONCLUSION: Gait speed tended to improve the predictive ability of FRAX more than OLST, but they both added value to FRAX.