Racial and Ethnic Differences in Total Knee Arthroplasty in the Veterans Affairs Healthcare System from 2001 - 2014.

Hausmann LR, Brandt C, Carroll C, Fenton BT, Ibrahim SA, Becker W, Burgess D, Wandner L, Bair MJ, Goulet JL. Racial and Ethnic Differences in Total Knee Arthroplasty in the Veterans Affairs Healthcare System from 2001 - 2014. Poster session presented at: VA HSR&D / QUERI National Meeting; 2015 Jul 9; Philadelphia, PA.

Related HSR&D Project(s)

• RCD 06-287 – Identifying Mechanisms Linking Perceived Discrimination & Health
• CRE 12-012 – Musculoskeletal Diagnoses Cohort: Examining Pain and Pain Care in the VA
• IIR 04-137 – Joint Replacement Utilization Disparity: The Role of Communication
• IIR 05-234 – Knee Replacement Disparity: A Randomized, Controlled Intervention

Objectives: Total knee arthroplasty (TKA) is an effective treatment that reduces pain and disability due to osteoarthritis (OA). Outside VHA, Blacks are less likely than Whites to undergo TKA, and the gap is increasing. We examined national trends in Black-White and Hispanic-White TKA differences within VHA using one of the largest Veteran cohorts in the nation. Methods: We used data from the HSRD-funded Musculoskeletal Disorders (MSD) Cohort, which includes over 5 million Veterans with MSD diagnoses who received VHA care between 2000-2014. For this analysis, we identified non-Hispanic White, non-Hispanic Black, and Hispanic Veterans 50 years or older with OA diagnoses (ICD-9 715.xx) recorded at 2 or more outpatient visits within 18 months or at least 1 inpatient visit from 2001-2011. Veterans were followed from their first OA diagnosis until 09/30/2014 to determine TKA receipt (CPT 27447 or ICD-9 81.54). We excluded those with TKA on or before their first OA diagnosis. We used Cox proportional hazard regression to examine effects of race/ethnicity, year of diagnosis, and race/ethnicity by year interaction adjusting for age, sex, pain intensity score on day of diagnosis, and medical and psychiatric comorbidities. Veterans who died or did not have TKA by 09/30/2014 were censored. Results: We identified 512,302 White, 53,869 Black, and 18,129 Hispanic Veterans with OA; TKA rates were 1.8%, 1.9%, and 2.4%, respectively. The interaction of race/ethnicity by year of diagnosis was significant (p = 0.04), and revealed that the Black-White Hazard Ratio increased over time, from 0.78 in 2002, to 0.64 in 2006, to 0.61 in 2009, to 0.59 in 2011. No significant differences were observed between Hispanic and White Veterans. Implications: In this large cohort, Black Veterans with OA were less likely to undergo TKA than Whites after controlling for demographic and clinical characteristics, with differences increasing over time. Impacts: This is the first study to follow a large national cohort of Veterans to examine trends in racial/ethnic differences over time. Unfortunately, Black-White differences in TKA have increased in VHA. Identifying, tracking, and understanding differences such as these is essential to fulfilling VHA's mission to deliver high-quality and equitable care to all Veterans.