The demographic characteristics of the study popula­tions are shown in Table 1. The mean age, age range and gender were comparable between the patient and control groups without statistically significant differences. The frequency of ApoE alleles in AD patients, non AD con­trols and the rest (others) are shown in Table 2. The fre­quency of the E4 allele in AD patients was increased al­most 2-fold (14.7%), relative to non AD controls (8.3%) and the rest (7.7%). Conversely, the frequency of E2 was reduced in AD patients (3.7%) relative to the other two groups (6.3 and 6.7%, respectively). These differences were significant at the p<0.05 level in the λ² test.

The distribution of ApoE genotypes in the three groups are shown in Table 3. The most frequent genotype in these groups was E3/E3. The most frequent genotype with at least one E4 allele in the AD population was E3/E4 (29.4%), compared to 15.7 and 12.4% in non AD controls and in the rest of the population, respectively. There were no individuals with E4/E4 or E2/E4 in the AD population. Testing for statistical significance was not possible because of null values in multiple cells.

The results of logistic regression analysis of putative risk factors for AD are shown in Table 4. The presence of at least one E4 allele was a significant risk factor for AD (OR = 2.21, 95% CI = 1.04-4.71; p = 0.04), whereas hav­ing a college education was associated with a significantly lower risk for AD (OR = 0.17, 95% CI = 0.04-0.66; p = 0.01).

Table 2. ApoE allele distribution in the populations studied

^a Probable AD + possible AD patients.

^b Non AD, cognitively normal individuals.

^c Subjects not clinically examined and those with a diagnosis other than AD.