Objectives. To examine patterns of use of end-of-life care in patients receiving treatment at a large, urban safety-net hospital from 2000 to 2010. Methods. Data from the Parkland Hospital palliative care database, which tracked all consults for this period, were analyzed. Logistic regression was used to identify predictors of hospice use, and Cox proportional hazards modeling to examine survival. Results. There were 5,083 palliative care consults over the study period. More patients were Black (41%) or White (31%), and younger than 65 years old (75%). Cancer patients or those who received palliative care services longer were more likely to receive hospice; those who had no form of health care assistance were less likely. There were no racial/ethnic differences in hospice use. Conclusion. In this cohort, there were no racial/ethnic disparities in hospice use. Those who had no form of health care assistance were less likely to receive hospice.