© 2015, Springer-Verlag Berlin Heidelberg. The aim of the study was to assess racial disparities in utilization rates and outcomes after primary total elbow arthroplasty (TEA). We used the National Inpatient Sample from 1998 to 2010, a US national database. Patient characteristics, comorbidity and outcomes after TEA were assessed over time and differences by race studied over the study period. Cochran–Armitage test was used for time trends and logistic regression for the comparison of outcomes by race. In 1998, TEA utilization rate was 0.38/100,000 in Whites and 0.24/100,000 in Blacks (p = 0.002); in 2010, it was 0.91 and 0.59/100,000, respectively (p < 0.0001). White–Black disparity in TEA utilization was significant across 13 years (p = 0.03). Compared with White patients, Black patients undergoing TEA were younger (61.9 vs. 52 years; p < 0.0001), less likely to be female (70.6 vs. 61.4 %; p = 0.0007) and more likely to have rheumatoid arthritis as the underlying diagnosis (13.0 vs. 17.2 %; p = 0.036). Mortality was rare, 0.26 % in Blacks and 0.32 % in Whites (p = 0.83). Discharge to an inpatient facility was higher in White versus Black patients in unadjusted analyses (16.8 vs. 10.4 %; p = 0.003), but in analyses adjusted for age, sex, Deyo–Charlson index and the underlying diagnosis, the differences were no longer significant (p = 0.79). The length of hospital stay greater than the median stay was noted in 29.8 % Whites versus 31.2 % Blacks, respectively (p = 0.61). There was no evidence of White–Black disparity in hospital length of stay in 1998–2000 (p = 0.66) or 2009–2010 (p = 0.59) periods. In this study, we found persisting racial disparities in TEA utilization rates. No White–Black disparities were noted in TEA outcomes, except slight differences in discharge disposition.