FRAX May Overestimate Fracture Risk in RA Patients

The 10-year risk of hip and major osteoporotic fracture is overestimated by the United Kingdom Fracture Risk Assessment Tool (FRAX) for patients with rheumatoid arthritis (RA) compared with the general population, according to research published in Annals of Rheumatic Disease. The FRAX tool currently does not take into account for potential adjustment factors such as disease severity or joint damage,

Corinne Klop, PharmD, from the Division of Pharmacoepidemiology & Clinical Pharmacology at the Utrecht Institute for Pharmaceutical Sciences, Utrecht, The Netherlands, and colleagues performed a cohort study of 625 practices in the United Kingdom using the Clinical Practice Research Datalink (CPRD). The researchers used the FRAX to assess performance with discrimination and calibration, later updating the methods to include recalibration and extension.

For differences in predictive performance, Dr Klop and colleagues used the UK National Osteoporosis Guideline Group Intervention thresholds, which included C-statistic and Net Reclassification Improvement (NRI) methods.

They analyzed data from 11,582 patients with RA and 38,755 patients from general practice. Of these, there were 7,221 patients with RA (297 hip fractures, 2,733 deaths) and 24,277 general practice patients (247 hip fractures, 1,699 deaths) linked to hospital admissions for hip fracture, according to CPRD Hospital Episode Statistics (CPRD-HES).

For RA patients in the CPRD, UK FRAX overestimated the risk of 10-year risk of hip fracture, with a mean predicted risk of 13.3% compared to a mean observed risk of 8.4% (95% CI, 7.8% – 9.0%). 

UK FRAX also overestimated the risk of major osteoporotic fracture in patients with RA and a mean predicted risk of 5.5% compared to a mean observed risk of 3.1% (95% CI, 2.8% – 3.5%). These results remained significant when linked to hospital data (5.5% vs. 4.1%; 95% CI, 3.6% – 4.6%) and when adjusting for antiosteoporotic drug (AOD) use (mean observed risk = 4.6%).

Regarding UK FRAX, the researchers reported good calibration for hip fracture in general population patients (2.7% vs. 2.4%), good discrimination for hip fracture in RA (0.78) and general population patients (0.83), and moderate discrimination for major osteoporotic fracture in RA (0.69) and general population (0.71) patients.

When extending the recalibrated UK FRAX model with factors such as RA disease duration, secondary osteoporosis, and glucocorticoids (> 7.5 mg per day), Dr Klop and colleagues found the extended model did not improve either the NRI (0.01; 95% CI, −0.04 – 0.05) or the C-statistic (0.78) regarding classification of hip fractures compared to non-cases.

Summary and Clinical Applicability

UK FRAX overestimated the risk for hip fracture and major osteoporotic fracture in patients with RA when fractures were measured in primary care data

Dr Klop and colleagues believed the overestimation of hip fracture risk by the FRAX model for patients with RA compared with the general population could be due to the increased mortality risk present in RA patients.

“Their lifespan is reduced by 3–10 years, for which no improvement has been found over the past decades,” Dr Klop and colleagues wrote. 

Limitations and Disclosures

Other factors such as exposure to disease-modifying anti-rheumatic drugs (DMARDs) and AOD treatment status may have also contributed to observed fracture risk. However, Klop and colleagues cited several studies that show FRAX appears to control for these factors.

“The finding that overestimation of UK FRAX for risk of hip fracture increased with longer duration of RA may relate to increased competing mortality with longer duration of RA disease, and greater loss of [bone mineral density] BMD during recent onset of disease,” Dr Klop and colleagues wrote.

References

Klop C, De vries F, Bijlsma JW, et al. Predicting the 10-year risk of hip and major osteoporotic fracture in rheumatoid arthritis and in the general population: an independent validation and update of UK FRAX without bone mineral density. Ann Rheum Dis. 2016; Published on March 16, 2016. doi:10.1136/annrheumdis-2015-208958.