The FRACTURE Index was derived from the Study of Osteoporotic Fractures (SOF), a prospective cohort of 9704 predominantly White, community-dwelling US women aged ≥65 years who were followed for about 5 years for incident hip, vertebral and other non-vertebral fractures. Black and colleagues used six easily reported clinical risk factors, with optional total hip bone mineral density (BMD), to create a simple 5-year fracture risk score for older postmenopausal women.

The score can be calculated using clinical factors alone or with added total hip T-score. In SOF, fracture risk rose steeply across quintiles of the FRACTURE Index, and the score was externally validated in the French EPIDOS study and incorporated into later guideline discussions on case-finding.

Study or setting	Population	Outcome and follow-up	Key findings for FRACTURE Index
SOF derivation (Black et al. 2001)	7,782 women from SOF with hip DXA (mean age ~73 years, almost all White, living in the community).	Hip, vertebral and non-vertebral fractures over ~5 years.	Across quintiles of the clinical FRACTURE Index (no BMD), 5-year hip fracture risk increased from ~0.6% to ~8.2 (≈14-fold difference). Adding total hip BMD gave similar or slightly higher gradients (hip risk ~0.4% to ~8.7) and improved prediction of vertebral and non-vertebral fractures. In the highest-risk quintiles, 5-year non-vertebral and vertebral fracture risks were about 26–28% and 10–11% respectively.
EPIDOS external validation	≈6,600–7,500 French women, generally ≥75 years, living in the community.	Hip fractures over about 4 years, extrapolated to 5-year hip fracture risk.	Applying the SOF-derived FRACTURE Index to EPIDOS showed good discrimination: estimated hip fracture risk remained low in the lowest quintile and was around 9–10% in the highest. Models with and without hip BMD produced large relative risk gradients between lowest and highest quintiles (approximately 23-fold with BMD and 6-fold without), although absolute risks differed between cohorts.
Guideline and case-finding use	Older community-dwelling postmenopausal women in guideline and health-system modelling work.	Short- to medium-term hip and other osteoporotic fractures.	National and international reports, including the US Surgeon General’s Bone Health report and WHO technical work, have used FRACTURE Index thresholds to inform case-finding. A clinical FRACTURE Index ≥4 (without BMD) has been suggested to select women for hip DXA, while combined scores ≥6 (with hip BMD included) highlight women at particularly high short-term fracture risk who warrant active prevention and treatment discussion.

Original derivation and most validation were in community-dwelling, predominantly White women aged ≥65 years. Calibration in men, younger postmenopausal women, other ethnic groups and residents of aged-care facilities is uncertain.
Women with major secondary osteoporosis (for example prolonged high-dose systemic glucocorticoids or other significant secondary causes) were largely excluded from derivation cohorts, so the score may underestimate risk in these groups.
The FRACTURE Index stratifies 5-year fracture risk rather than providing a 10-year absolute probability. Contemporary guidelines often prefer FRAX or similar 10-year tools for treatment thresholds, but the FRACTURE Index remains a useful, simple checklist where hip BMD or FRAX are not readily available.
Event rates and optimal cut-offs vary by population. Interpret the score alongside BMD, prior minimal-trauma fractures, falls risk and local osteoporosis pathways rather than as a stand-alone treatment rule.

FRACTURE Index 5-year hip fracture risk score