Background/Purpose:  Rheumatoid arthritis (RA) is a well-known risk factor for osteoporosis and hip fracture. Recent studies suggest RA patients fracture at a younger age and suffer higher rates of morbidity and mortality[1]. The purpose of this study is to characterize RA patients with hip fracture in an integrated healthcare system in the US.

Methods:  This retrospective cohort study, conducted in Kaiser Permanente Northern California, examined data from 13,550 women ≥ age 65 years with an incident hip fracture during 2000-2010. Demographic characteristics, comorbidity index (Charlson, CCI), prior fracture history and recent bisphosphonate (BP) therapy (2 prescriptions £1 year prior to hip fracture) were determined using health plan databases. RA was defined by a problem-list diagnosis and ≥3 visit diagnoses of RA. Rehospitalization (£30 days of discharge) and mortality outcome (£1 year) were assessed post-fracture. Standard descriptive statistics were used to examine differences in age, race/ethnicity, CCI, recent BP use, and prior fractures among women with and without RA experiencing hip fracture. Multivariable logistic regression analyses were used to examine the association of RA and mortality, rehospitalization, and recent BP use.

Results:  Among 13,550 women who had a hip fracture, 339 (2.5%) had RA. Women with RA were slightly younger compared to women without RA (mean age 79.4±6.9 vs 82.5±7.4), and were twice as likely to be under age 75 (29.5 vs 15.1%, p<0.01). A larger proportion of RA patients were of non-white race/ethnicity (23.3 vs 16.3%), had greater comorbidity (CCI≥3, 37.5 vs 21.5 %), and were more likely to have had a prior fracture (44.8 vs 37.3%; all p<0.01). Fracture type (femoral neck vs trochanteric fracture) was similar between the two groups. Overall mortality rates at 1, 3, 6, and 12 months did not differ significantly for women with vs without RA (5.0 vs 6.4%, 10.6 vs 12.8%, 14.8 vs 17.0%, and 19.8 vs 22.9%, respectively), and RA status was not associated with greater mortality outcome even after adjusting for differences in age, race/ethnicity, prior fracture, recent BP use and CCI (adjusted odds ratio, OR 0.9, 95% CI 0.7-1.2). Readmission rate within 30 days was also similar for women with and without RA (12.7 vs 12.0%, p=0.69), with no increased risk for women with RA (adjusted OR 1.0, CI 0.7-1.4). However, women with RA had 3-fold greater odds of having received bisphosphonate therapy within the year prior to hip fracture (adjusted OR 3.0, CI 2.3-3.8).

Conclusion: Women with RA were younger and had greater comorbidity at the time of hip fracture. However, RA status did not appear to be independently associated with increased morbidity and mortality post hip fracture. The higher proportion of RA women with a prior fracture and evidence of recent BP therapy is consistent with their higher underlying fracture risk. Future studies should focus on prevention strategies to decrease risk of hip fracture in RA patients.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/outcomes-of-rheumatoid-arthritis-patients-with-hip-fracture/