Published studies of the prognostic value of left ventricular (LV) hypertrophy and LV geometric pattern in African-Americans were based on referred or hospitalized patients with hypertension or coronary heart disease. All-cause mortality rates and survival associated with LV geometric pattern were determined using echocardiography in a population-based sample of middle-aged and elderly African-American men and women. During the third (1993 to 1995) visit of the ARIC Study, echocardiography was performed at the Jackson, Mississippi, field center on the cohort of 2,445 African-Americans aged 49 to 75 years. M-Mode LV echocardiographic measurements were available for 1,722 persons. Mortality data were available through December 31, 2003. During the follow-up period (median 8.8 years, maximum 10.4), 160 deaths were identified. In men, multivariable-adjusted hazard ratios for all-cause mortality (compared with men with normal LV geometry) were 1.75 (95% confidence interval [CI] 0.71 to 4.33) in those with concentric LV hypertrophy, 0.38 (95% CI 0.08 to 1.88) in those with eccentric LV hypertrophy, and 0.79 (95% CI 0.41 to 1.54) in those with concentric remodeling. In women, multivariable-adjusted hazard ratios for all-cause mortality (compared with women with normal LV geometry) were 1.17 (95% CI 0.48 to 2.84) in those with concentric LV hypertrophy, 1.23 (95% CI 0.46 to 3.28) in those with eccentric LV hypertrophy, and 1.17 (95% CI 0.60 to 2.28) in those with concentric remodeling. In conclusion, in this population-based cohort of middle-aged and elderly African-Americans free of coronary heart disease, adjustment for baseline differences in cardiovascular disease risk factors and LV mass greatly attenuated the strength of the association between LV pattern and all-cause mortality risk in women. In men, an association between concentric LV hypertrophy and mortality risk remained.