The aim of this study was to examine the effectiveness of implantable cardioverter-defibrillator (ICD) therapy in whites and blacks who were enrolled in the Multicenter Automatic Defibrillator Implantation Trial-II (MADIT-II). The study population involved 1,232 subjects, with 1,073 white patients (87%) and 102 black patients (8%). Univariate analyses revealed a significant reduction in mortality in white patients treated with ICD therapy (p <0.02), but not in black patients (p = 0.96). White patients had a significant reduction of sudden cardiac death (p <0.01), but no benefit was evident in black patients (p = 0.62). Adjusting for relevant covariates, the ICD therapy/conventional therapy hazard ratios for total mortality were favorable in whites (0.75, 95% confidence interval [CI] 0.55 to 1.02), but not in blacks (1.25, 95% CI 0.42 to 3.60); the hazard ratios for sudden cardiac death were significantly (p = 0.04) lower in whites (0.29, 95% CI 0.17 to 0.49) than in blacks (1.71, 95% CI 0.33 to 8.84). Interaction analyses revealed a pattern of progressive improvement in ICD efficacy in whites as the end point became more specific for arrhythmic death, with exactly the reverse for blacks. In conclusion, ICD therapy in MADIT-II was associated with a reduction in total mortality, cardiac death, and sudden cardiac death in whites but not in blacks.