Objective: To examine the influence of the proband's APOE genotype on AD am
ong first-degree relatives in a community-based study of African Americans,
whites, and Caribbean Hispanics. Methods: History of AD and demographic in
formation were obtained on 1,073 siblings and parents of 312 patients with
AD and 2,722 siblings and parents of 802 nondemented controls. APOE genotyp
ing was performed on all 1,114 patients and controls. Results: A higher pro
portion of patients with AD (35%) than controls (27%) had one or more APOE-
epsilon 4 alleles (p = 0.03). When compared with relatives of controls with
out an APOE-epsilon 4 allele, the risk for AD was increased in first-degree
relatives of both patients (rate ratio [RR] = 1.9, 95% confidence interval
[CI] = 1.2 to 3.1) and controls (RR = 1.8, 95% CI = 1.2 to 2.6) with one o
r more APOE-epsilon e alleles, regardless of ethnic group. There was a simi
lar trend of increased risk in relatives of patients without an APOE-epsilo
n 4 allele, but this was limited to Hispanics and African Americans. Conclu
sions: The presence of an APOE-epsilon 4 allele increases risk for AD among
first-degree relatives, regardless of the probands' disease status, among
all ethnic groups. Relatives of patients without an APOE-epsilon 4 allele w
ere also at increased risk for AD among Hispanics and African Americans, su
ggesting that other genes or risk factors may influence risk.