Use of FRAX as a Determinant for Risk-Based Osteoporosis Screening May Decrease Unnecessary Testing While Improving the Odds of Identifying Treatment Candidates

Purpose: We have assessed the hypothetical impact of guideline-concordant osteoporosis screening on baseline behaviors utilizing two different guidelines and determined the relative ability of each to identify osteoporosis treatment candidates. Methods: We conducted secondary analyses from the Fracture Risk Perception Study, which enrolled patients aged 50 to 75years to complete questionnaires about their bone health. We determined our baseline screening rates and detection of treatment candidates and then assessed the hypothetical impact of adherence to U.S. Preventive Services Task Force (USPSTF) and National Osteoporosis Foundation (NOF) criteria, particularly for women aged 50 to64. Results: Of 144 women aged 50 to 64years screened, 14 (9.7%) were treatment candidates. Screening based on identification of one or more risks (NOF) would lead to testing of 102 of the 144 patients (71%) to identify 12 of 14 treatment candidates (86%). Applying USPSTF criteria (9.3% FRAX threshold) would test 45 of the same 144 women (31%) to identify 11 of 14 treatment candidates (79%). NOF risk-based criteria would result in a moderate absolute screening rate reduction (16%, p=.0011; 95% CI, 7%-25%), but only marginal improvement in identifying treatment candidates (odds ratio, 2.67; 95% CI,0.57-12.47). Applying the more selective USPSTF criteria greatly reduced unnecessary testing (56% absolute screening rate reduction; p<.0001; 95% CI, 47%-64%) while further improving the odds of identifying treatment candidates (odds ratio, 10.35; 95% CI,2.72-39.35). Conclusions: When contemplating screening younger patients, systematic calculation of FRAX and ordering only when the 9.3% fracture risk threshold is reached may decrease unnecessary screening for many women while still identifying appropriate osteoporosis treatment candidates.