Fracture Risk Assessment in Persons with Rheumatoid Arthritis

Rheumatoid arthritis (RA) is a central risk factor for osteoporosis. Persons with RA have up to two times higher risk of osteoporotic fracture, which are associated with significant morbidity and excess mortality. Despite the burden of fractures in RA, many persons with RA at high risk of fracture are not taking one of several widely available pharmacologic therapies to treat osteoporosis. Concerns about rare side effects (i.e. atypical femoral fractures, osteonecrosis of the jaw) of bone heath medications, particularly bisphosphonates, are prevalent.The most commonly used fracture risk assessment tool in persons with RA is FRAXTM, which produces 10-year risk probabilities of major osteoporotic fracture (MOF; hip, clinical spine, forearm or humerus fracture) and hip fracture. When shared decision-making discussions regarding bone health medications are framed in terms of long-term (i.e. 10-year) fracture risk, persons often perceive that potential medication side effects or other health problems take precedence over possible fracture prevention over such a long horizon. However, fracture risk is not stable over time. Imminent (i.e. 2-year) fracture risk may be of greater immediate personal significance to persons with RA considering initiation of pharmacologic therapy for fracture prevention.We hypothesize that we can determine how well the 10-year MOF and hip fracture risk probabilities generated from U.S. FRAX capture imminent MOF and hip fracture risk in persons with RA age 65 years and older. We will use the Rheumatology Informatics System for Effectiveness (RISE) registry in conjunction with linked Medicare claims data to assess the discriminative ability of U.S. FRAX to predict incident imminent and intermediate horizon (i.e. 4-year) risk of MOF and hip fracture.We further hypothesize that imminent fracture risk prognostication in persons with RA can be improved by incorporating RA-specific characteristics (i.e. seropositivity for rheumatoid factor (RF) and/or anti-cyclic citrullinated peptide antibody (ACPA), RA disease activity level, and RA therapy). To assess this, we will determine if the addition of RA-specific characteristics to models already incorporating the individual risk factors in U.S. FRAX improves model predictive performance for imminent and intermediate horizon risk of MOF and hip fractures in our older RA cohort from RISE.