Comparing self-perceived and estimated fracture risk by FRAX(R) of women with osteoporosis

Abstract

In this study, we compared subjective fracture risks of Hungarian women with osteoporosis to FRAX(R)-based estimates. Patients with a previous fracture, parental hip fracture, low femoral T-score, higher age, and higher BMI were more likely to underestimate their risks. Patients also failed to associate risk factors with an increased risk of fractures. PURPOSE: The main objectives were to explore associations between self-perceived 10-year fracture risks of women with osteoporosis (OP) and their risks calculated by the FRAX(R) algorithm and to identify determinants of the underestimation of risk. METHODS: We carried out a cross-sectional study in 11 OP centers in Hungary and collected data on the risk factors considered by the FRAX(R) calculator. Patients estimated their subjective 10-year probability of any major osteoporotic and hip fracture numerically, in percentages and also on a visual analog scale (VAS). We compared subjective and FRAX(R) estimates and applied logistic regression to analyze the determinants of the underestimation of risk. Associations between risk factors and subjective risk were explored using linear probability models. RESULTS: Nine hundred seventy-two OP patients were included in the analysis. Major OP and hip fracture risk by FRAX(R) were on average 20.1 and 10.5%, while subjective estimates were significantly higher, 30.0 and 24.7%, respectively. Correlations between FRAX(R) and subjective measures were very weak (r = 0.12-0.16). Underestimation of major OP fracture risk was associated with having had a single previous fracture (OR = 2.0), parental hip fracture (OR = 3.4), femoral T-score </=-2.5 (OR = 4.2), higher age, body mass index, and better general health state. We did not find significant associations between subjective risk estimates and most of the risk factors except for previous fractures. CONCLUSIONS: Hungarian OP patients fail to recognize most of the risk factors of fractures. Thus, education of patients about these risk factors would be beneficial especially for the elderly with a low femoral T-score and parental hip fracture history.