Original articles

The application of FRAX® to determine intervention thresholds in osteoporosis treatment in Poland

Janusz E. Badurski, John A. Kanis, Helena Johansson, Andrzej Dobreńko, Nonna A. Nowak, Stefan Daniluk, Elżbieta Jezienicka
Published online: May 01, 2011

INTRODUCTION: FRAX®, an assessment algorithm for estimating fracture probability, has been widely used for the evaluation of osteoporosis since 2008. Its clinical use requires that osteoporotic fracture probability is established at which treatment can be recommended. OBJECTIVES: The aim of the present study was to explore possible treatment thresholds for Poland. PATIENTS AND METHODS: The FRAX‑based probabilities were calculated in 1608 unselected postmenopausal women from Białystok using the British model (version 3.1). Intervention thresholds were set at fracture probability equal to women with a bone mass density (BMD) T‑score of –2.5 standard deviation (criterion A), equal to women with a prior fracture using a fixed threshold irrespective of age (criterion B), or an age‑dependent threshold (criterion C). Additionally, we assumed that all women with a prior fracture would be eligible for treatment plus those with a fracture probability equal to women with a prior fracture using a fixed threshold (criterion D). RESULTS: Mean 10‑year probability of a major osteoporotic fracture was 10.9% when BMD was not included in the FRAX calculation and 11.6% with BMD included. In women with a prior fragility fracture, the respective probabilities were 18.0% and 17.4%. For criterion A, 39% women aged 50 years or more would be eligible for treatment, for criterion D – 35%, and for criteria B and C – 16%. For criteria B and C, women with higher risk would be eligible for treatment compared with criteria A and D. Assuming a relative fracture risk reduction of 30%, the number needed to treat (NNT) to prevent a major fracture was lower for criteria B and C (NNT = 13 and 14, respectively) than for criteria A and D (NNT = 18). CONCLUSIONS: The use of intervention thresholds based on the probabilities equal to women with a prior fracture is most efficient. The use of an age‑specific threshold may be more clinically appropriate than a fixed probability threshold for all ages.

Full-text article available only as a pdf file for download

Download article