We aimed to determine the association and related factors of the apolipoprotein E (ApoE) genotype and Alzheimer's disease (AD) in Taiwan. We examined ApoE genotypes in 50 Chinese patients with AD and 50 age- and sex-matched controls. The patients met the criteria of probable AD of the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) and AD of the Diagnostic and Statistical Manual of Mental Disorders 4th edition (DSM-IV). There were 28 females and 22 males in the case and control groups. The mean age of onset of AD was 72. 62 years. The average interval between onset and research was 3.85 years. The frequency of ApoE epsilon 4 in the AD group was significantly higher than that in the controls (0.13 versus 0.02, p < 0.05). The odds ratio for AD in individuals with at least one ApoE epsilon 4 allele was 6.0 (95% CI 1.34 to 55.3, p < 0.001). The linear trend for AD in proportion to alleles of ApoE epsilon 4 was significant (chi 2 = 8.3, p = 0.004). The risk of ApoE epsilon 4 allele for the late-onset AD patients, males, or those who received less education was higher than that for the early-onset AD patients, females, or those who had received more education. The sensitivity of the epsilon 4 allele was 24%, the specificity 96%, the positive predictive value 86%, and the negative predictive value 56%. Our results supported that the ApoE epsilon 4 allele is related to AD in Taiwan. In addition, sex and education may play important roles in the presence of ApoE epsilon 4 allele. The epsilon 4 allele seemed helpful as an adjunct for diagnostic testing of AD.