What is the FRAX Score?
FRAX (Fracture Risk Assessment Tool) is a WHO-developed online calculator that estimates a patient's 10-year probability of hip fracture and major osteoporotic fracture using easily obtainable clinical information, with or without bone mineral density, to guide treatment decisions in adults over 50. 1
Core Components and Calculation
FRAX calculates fracture risk using the following clinical inputs 1, 2:
- Patient demographics: Age (40-90 years), sex, race/ethnicity, body mass index, height, and weight 1, 3
- Clinical risk factors: Prior fragility fracture, parental hip fracture history, current tobacco smoking, alcohol consumption (≥3 units/day), glucocorticoid use, rheumatoid arthritis, and secondary causes of osteoporosis 1, 2
- Optional BMD input: Femoral neck T-score from DXA (not required but improves accuracy) 1
The tool generates two separate risk estimates 4, 5:
- 10-year probability of hip fracture
- 10-year probability of major osteoporotic fracture (clinical vertebral, hip, forearm, or humerus fractures)
Treatment Thresholds
Pharmacologic treatment is recommended when FRAX scores indicate ≥3% 10-year hip fracture risk OR ≥20% 10-year major osteoporotic fracture risk. 1, 4, 2
For postmenopausal women aged 50-64 years, screening is indicated when FRAX scores (calculated without BMD) approach or exceed the baseline risk of a 65-year-old white woman with no risk factors (9.3% 10-year major fracture risk, 1.3% hip fracture risk) 1, 4.
Risk Stratification for Treatment Selection
Very high-risk patients (major osteoporotic fracture >30% OR hip fracture >4.5%) should be considered for anabolic therapy first (teriparatide, abaloparatide, romosozumab) 4, 2.
High-risk patients (major osteoporotic fracture ≥20% OR hip fracture ≥3%) should be considered for antiresorptive therapy such as bisphosphonates or denosumab 4, 2.
Critical Adjustments for Glucocorticoid Users
For patients taking prednisone >7.5 mg/day, manually adjust the calculated FRAX scores by multiplying major osteoporotic fracture risk by 1.15 and hip fracture risk by 1.2. 4, 2 This adjustment is necessary because FRAX only uses binary (yes/no) glucocorticoid exposure and does not account for dose-dependent effects 4, 2.
Predictive Accuracy
FRAX demonstrates moderate-to-good predictive accuracy with area under the curve values of 0.74-0.79 for hip fracture prediction and 0.67-0.71 for major osteoporotic fracture prediction when BMD is included 4. Age and femoral neck T-score are the strongest contributors to fracture risk, while BMI has marginal contribution 6. Among clinical risk factors, parental hip fracture history and ongoing glucocorticoid treatment have the largest additive effects 6.
Clinical Utility and Screening Benefits
Screening with FRAX followed by DXA for high-risk individuals reduces hip fractures by 17% (RR 0.83,95% CI 0.73-0.93) and major osteoporotic fractures by 6% (RR 0.94,95% CI 0.88-0.99) over 3.7-5 years 4. Treatment with bisphosphonates in high-risk patients reduces vertebral fractures by 49% (RR 0.51,95% CI 0.39-0.66) and hip fractures by 33% (RR 0.67,95% CI 0.45-1.00) 4.
Reassessment Intervals
Repeat FRAX assessment every 1-3 years for patients on glucocorticoids not receiving osteoporosis treatment, with earlier reassessment for very high-dose users 4, 2. For patients with osteopenia and low initial FRAX scores, repeat in 2 years, or in 1 year if new risk factors develop 2.
Important Limitations and Pitfalls
FRAX is only validated for untreated patients aged 40-90 years and should not be used to assess fracture risk reduction in individuals already on treatment, though it may still guide decisions about continued treatment 7, 3.
FRAX does not account for several important risk factors 4, 2:
- Dose-dependent effects of glucocorticoids beyond yes/no input
- Fall history or frailty status
- Lumbar spine BMD or trabecular bone score
- Number of prior fractures (only captures presence/absence)
- Diabetes mellitus
Race-specific calculators systematically predict lower fracture risk for Asian, Black, and Hispanic individuals compared to White individuals with identical clinical profiles, potentially leading to treatment disparities. 4, 2 This is a critical equity concern that clinicians must recognize when applying FRAX across diverse populations.
For women 65 years and older with a previous fracture, 98% of FRAX combinations exceed the treatment threshold regardless of T-score or other factors 6. For women age 70+ with a parental hip fracture history, 99% of FRAX combinations exceed the treatment threshold 6.
Practical Access
FRAX is freely accessible online at www.shef.ac.uk/FRAX/ and has been incorporated into some DXA reporting software 1, 5. The tool is country-specific and has been validated in multiple large U.S. cohorts 1.