We explored the impact of substance dependence on the efficacy of an HIV sexual risk reduction intervention previously shown to be effective among men with severe mental illness by comparing rates of high-risk sexual behaviors among men with (n = 26) and without (n = 31) a lifetime history of substance dependence. We sub-divided subjects by alcohol and drug dependence status, comparing each intervention sub-group to the corresponding control sub-group. At each follow-up interval (six, 12 and 18 months), the intervention group as a whole and the non-substance dependent participants showed a significant reduction in risk; the substance-dependent men showed no difference from controls. These data suggest that among men with severe mental illness, substance dependence may be a further impediment to HIV risk reduction.