The Alzheimer's Association and the National Institute on Aging sponsored a meeting of experts in Alzheimer's disease (AD), geneticists, social scientists, and ethicists in Chicago in October 1995 to discuss the use of apolipoprotein E (APOE) genotyping in Alzheimer's disease. A short scientific report was published in the scientific journal Lancet with recommendations from the group. Several areas were discussed, including: (1) the scientific basis for recommendations on the application and uses of APOE genotyping, (2) clarifying the clinical and epidemiological research that needs to done, (3) genetic counseling issues, (4) ethical and legal issues, and (5) potential uses of APOE genotyping for treatment care planning. This contribution was a general introduction to begin the meeting. The genetic association of APOE genotypes with the age of onset distribution and risk of Alzheimer's disease was reviewed. An analysis of the current applications for three distinctly different applications of APOE genotyping was presented with the following conclusions: (1) predictive testing for cognitively intact persons was not recommended; (2) APOE genotyping is a promising adjunct for use in the differential diagnosis of patients with dementia; and (3) APOE genotyping may have a use in selecting therapies; however, further prospective studies are necessary. There is no universal "APOE test for AD." A strong emphasis was made to avoid use of the term in making recommendations regarding APOE genotyping without specific reference to the type of application involved. The predictive testing of asymptomatic persons versus APOE genotyping as a diagnostic adjunct for symptomatic patients has been seriously confused in both the lay and clinical press. The former application is not recommended, but diagnostic usefulness early in clinical evaluations for dementia has been confirmed.