OR10-2 Fracture Among Older and Younger HIV+ Medicare Beneficiaries

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: OR10-Osteoporosis
Saturday, June 15, 2013: 11:30 AM-1:00 PM
Presentation Start Time: 11:45 AM
Room 121 (Moscone Center)
Amy H Warriner*1, Wilson Smith2, Jeffrey R Curtis2, Kenneth G Saag1 and Elizabeth Delzell2
1Univ of Alabama at Birmingham, Birmingham, AL, 2University of Alabama at Birmingham
Background: Bone mineral density is lower in people with HIV (HIV+) compared to HIV-negative (HIV-) persons. However, it is still unclear if these changes convey an increased fracture risk among HIV+ persons. Use of a large administrative database, such as Medicare, allows for assessment of a large, diverse population.

Methods: Retrospective cohort study using Medicare beneficiaries and their medical claims from 1999-2010. From the national random 5% sample of Medicare beneficiaries, we selected subjects of any age; continuously enrolled for ≥12 months in Medicare parts A and B and not in a Medicare Advantage plan. All patients had a 12-month baseline period, after which follow-up began and continued until the earliest of fracture, loss of Medicare coverage, death, or 12/31/2010. We classified beneficiaries as HIV+ or HIV- based on claims during the baseline and follow-up periods using a previously developed HIV case-finding algorithm. We identified fracture outcomes during follow-up using claims data on diagnoses and procedures and included all fracture types (hip, spine, femur, pelvis, tibia/fibula, ankle, clavicle, humerus, radius/ulna, carpal bones, foot, rib). We used multivariable Poisson regression to estimate the rate ratio (RR) for the association between HIV status and fracture, adjusting for demographic factors and comorbidities assessed during baseline. Fracture occurrences at individual fracture sites were determined for HIV+ and HIV- groups.

Results:  The overall study population included 13,221 HIV+ and 2,500,442 HIV- beneficiaries.  The HIV+ population was more likely to be African-American, male, and younger, compared to the HIV- population. Crude fracture rates were 13.8 per 1000 person years in HIV+ patients vs. 20.5 per 1000 person years in HIV-patients. However, the RR of fracture was 1.67 (95% Confidence Interval, CI, 1.55-1.80) for HIV+ compared to HIV- subjects after adjustment for age, gender, year, geographic region, race, Medicare entry reason, Medicare/Medicaid dual-eligibility, hepatitic C and Charlson score.  The elevated association between HIV and fracture risk was present in both younger beneficiaries (<65 years old, RR 1.32, 95% CI 1.21 – 1.45) and older beneficiaries (≥65 years old, RR 1.52 (1.34 – 1.73) after adjustment. The most common fractures in younger HIV+ patients were hip, vertebral, ankle, and wrist.

Conclusions:  The risk of fracture is approximately 50% higher in HIV+ Medicare patients when compared to a HIV- group of Medicare patients after adjusting for demographics and other fracture risk factors.  The risk is present for both older and younger HIV+ persons but common “osteoporotic” fractures may occur at a younger age in HIV+ patients.

Nothing to Disclose: AHW, WS, JRC, KGS, ED

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm

Sources of Research Support: AHRQ CER K12 funding (1K12HS019465-02)