OBJECTIVES: To evaluate the association between apolipoprotein E (apo E) epsilon4 and mortality, the population attributable risk for mortality with epsilon4, and relative contributions of cardiovascular disease (CVD) and Alzheimer's disease (AD). DESIGN: Population-based cohort study. SETTING: Community-based. PARTICIPANTS: Permanent residents of Cache County, Utah, aged 65 and older as of January 1, 1995. MEASUREMENTS: Participants were genotyped at the apo E locus using buccal-swab deoxyribonucleic acid. Cardiovascular health was ascertained using self- or proxy-report interviews at participants' residences. AD was diagnosed according to Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised, and National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders criteria. Utah Department of Vital Statistics quarterly reports were reviewed to identify participants who died. RESULTS: Crude evaluations showed nonsignificantly greater risk of death for epsilon2/2 (hazard ratio (HR)=1.66, 95% confidence interval (CI)=0.92-2.76) and epsilon3/4 (HR=1.11, 95% CI=0.97-1.26) genotypes and significantly greater risk for epsilon4/4 (HR=1.48, 95% CI=1.09-1.96). After adjustment for age, age(2), sex, and education, risks increased to 1.98 (95% CI=1.08-3.35), 1.28 (95% CI=1.12-1.46), and 2.02 (95% CI=1.47-2.71), respectively, compared with epsilon3/3 genotypes. Adjustment for presence of any CVD did not change the risk of death for epsilon3/4 and epsilon4/4. Adjustment for AD reduced the risk of death for epsilon3/4 (HR=1.13, 95% CI=0.99-1.30) and epsilon4/4 (HR=1.59, 95% CI=1.15-2.14). The population attributable risk of death for epsilon3/4 and epsilon4/4 genotypes combined is estimated at 9.6%. CONCLUSION: These findings suggested that the epsilon2/2, epsilon3/4, and epsilon4/4 genotypes have greater early mortality risks. Further analyses showed that AD partially mediates the association between epsilon3/4, epsilon4/4, and death.