What is the significance of the Frax (Fracture Risk Assessment) score in clinical practice?

FRAX Score Significance in Clinical Practice

FRAX is a validated WHO algorithm that calculates 10-year absolute fracture probability and serves as the primary tool for identifying which patients with osteopenia or clinical risk factors should receive pharmacologic osteoporosis treatment, fundamentally improving upon BMD-alone approaches that miss the majority of patients who will fracture. 1

Core Clinical Utility

FRAX addresses the critical limitation that most fractures occur in patients with BMD T-scores above -2.5, making BMD alone insufficiently sensitive for fracture prediction despite its specificity. 1 The algorithm integrates easily obtainable clinical risk factors—age, BMI, prior fragility fracture, parental hip fracture history, smoking, glucocorticoid use, rheumatoid arthritis, secondary osteoporosis causes, and excessive alcohol—with optional femoral neck BMD to generate 10-year probabilities for major osteoporotic fracture (hip, clinical spine, humerus, or wrist) and hip fracture specifically. 1

Treatment Decision Thresholds

Pharmacologic treatment is recommended when FRAX demonstrates ≥3% 10-year hip fracture risk or ≥20% 10-year major osteoporotic fracture risk. 2, 3 These thresholds identify the subset of patients with osteopenia who warrant intervention, addressing the public health reality that over half of osteoporotic fractures occur in this larger population segment rather than in those with frank osteoporosis. 4

For screening postmenopausal women aged 50-64 years, the USPSTF recommends using a threshold equivalent to a 65-year-old white woman's baseline risk (9.3% 10-year major fracture probability), though this can be adjusted based on individual patient values and menopausal status. 1

Age-Dependent Intervention Approach

European guidance (ESCEO/International Osteoporosis Foundation) and UK National Osteoporosis Guideline Group recommend age-specific intervention thresholds that rise with age, set at the probability conveyed by a prior fragility fracture at that age (without considering BMD or other risk factors, assuming average BMI). 1 This approach provides equity across age groups and accounts for the reality that FRAX integrates mortality risk with fracture risk, effectively representing lifetime fracture probability in older adults with <10-year life expectancy. 1

Risk Stratification for Treatment Selection

Patients should be categorized as "high risk" versus "very high risk" to guide whether antiresorptive therapy (bisphosphonates, denosumab) or more expensive anabolic therapy (teriparatide, romosozumab) is initiated. 1 Very high risk is defined as fracture probability above 1.2 times the intervention threshold, or alternatively as ≥30% 10-year major osteoporotic fracture risk or ≥4.5% hip fracture risk. 1

Key Contributors to Very High Risk:

• Recent fracture (within 24 months): Risk is acutely elevated and wanes over 2 years but never returns to baseline; 31-45% of recurrent fractures occur within 1 year of sentinel fracture. 1
• High-dose glucocorticoids: ≥30 mg/day prednisone for >30 days or cumulative ≥5g/year. 1
• Multiple risk factor combinations, particularly older age with recent fracture and glucocorticoid use. 1

Glucocorticoid-Specific Adjustments

For patients on glucocorticoids >7.5 mg/day prednisone equivalent, multiply the standard FRAX major osteoporotic fracture risk by 1.15 and hip fracture risk by 1.2 to account for dose-dependent effects not captured in the base algorithm. 1, 2 FRAX with this adjustment should be calculated within 6 months of initiating chronic glucocorticoid therapy (≥2.5 mg/day for >3 months) in adults ≥40 years. 1, 2

BMD Integration Considerations

Including femoral neck BMD significantly improves FRAX predictive accuracy and should be incorporated when available. 2, 5 However, FRAX calculated without BMD may inappropriately recommend treatment for older patients with normal T-scores (where age alone drives high fracture probability) or fail to recommend treatment for younger patients with high BMI and low T-scores (where BMD would reveal significant risk). 5 This creates two distinct problematic scenarios affecting 10.6% of cases in one validation study. 5

Reassessment Intervals

For patients on continued glucocorticoids at low or moderate fracture risk who are not receiving osteoporosis therapy, reassess FRAX every 1-2 years. 1, 2 For those on osteoporosis treatment, reassess every 1-2 years to determine if BMD is stable, improving, or declining, which may warrant therapy changes. 1 Earlier reassessment is appropriate for very high-dose glucocorticoid users or those with incident fractures. 2

Critical Limitations and Pitfalls

FRAX is validated only for untreated patients aged 40-90 years and should not be used in adults <40 years, children, or those already on osteoporosis therapy. 1, 2, 4 For younger adults <40 years on glucocorticoids, use clinical risk assessment with BMD Z-scores instead; moderate risk is defined as Z-score <-3 or ≥10% bone loss over 1 year while on ≥7.5 mg/day prednisone for ≥6 months. 1

FRAX does not account for dose-response relationships beyond the glucocorticoid adjustment described above, particularly for alcohol consumption. 2 It also has race-specific calculation limitations that may create disparities in treatment recommendations among otherwise identical patients. 2

Falls risk, number of prior fractures, and fracture site are not captured in standard FRAX calculations, though FRAXplus now incorporates multipliers for recent fractures and specific fracture sites. 1, 2 Clinicians must use judgment to incorporate fall history and other factors not represented in the algorithm. 4

Comparative Performance

FRAX demonstrates incremental improvement over simplified tools like the Canadian CAROC system, with net reclassification improvement of +2.8% overall (P<0.001), particularly benefiting those who remain fracture-free. 6 Only 36 individuals need assessment with FRAX instead of simplified tools to yield one improved prediction (8 for prior fracture patients, 4 for prolonged glucocorticoid users). 6

In contemporary rheumatoid arthritis patients, FRAX remains well-calibrated except for slight overestimation in those with predicted risk >20%, supporting its continued use in this population despite predating modern RA treatment. 7