Objective: In a previous study of outpatients with chronic mental illness who were treated with neuroleptic medication, the authors found a higher incidence of tardive dyskinesia among blacks than among whites. in this study the authors examined psychosocial, clinical, treatment, and medical correlates of race that might explain this finding. Methods: Extensive baseline data were obtained for 398 outpatients at risk for tardive dyskinesia. Pearson and Mantel-Haenszel chi square analyses were used to determine significant associations between a large number of variables and race. Results and discussion: Compared with whites, nonwhites were more likely to be younger, less skilled, and unmarried; to report contacts with relatives outside the home; to have a diagnosis of schizopbrenia; and to receive higher doses of neuroleptic drugs primarily through depot medications. Non-whites were less likely to receive nonneuroleptic psychotropics, to wear dentures, and to report social contact with fathers. None of these racial differences explained more than a trivial portion of the association between race and the incidence of tardive dyskinesia. Conclusions: These results suggest the need for studies of differences in diagnostic and prescribing practices for black and white patients and of differences in the action and side effects of neuroleptics.