A multi-site descriptive outcome study examined differences between black and white veterans in admission characteristics, program participation, and outcomes following an episode of treatment in a Veterans Affairs residential program for homeless veterans with psychiatric and substance abuse problems. Admission, discharge, and 6-month and 12-month postdischarge follow-up data were collected on 119 black and 144 white veterans admitted to the Domiciliary Care for Homeless Veterans Program at three sites. Chi-square tests and t-tests were used to identify differences between racial groups at admission. Factorial repeated-measures analysis of covariance was then used to identify differences between black and white veterans in improvement, controlling for differences that were significant at the time of admission. On admission, blacks were younger and had more problems with drugs and violent behavior, but were less likely than whites to have clinical diagnoses of alcohol abuse or a serious psychiatric disorder, and had fewer suicide attempts. They also had more social contacts and had more frequently experienced a recent disruption in an important relationship. Few differences were found between the two racial groups in measures of program participation. One year after discharge, both black and white veterans had improved in virtually all domains. Black veterans showed greater improvement in medical symptomatology, social contacts, and violence, while white veterans showed a greater increase in outpatient health service use. While both black and white veterans benefitted from participation in residential treatment, the data suggest that blacks were more likely to re-establish previously disrupted social ties while whites increased their involvement in the VA health care system.