A case-control study is proposed to investigate the hypothesis that aluminum is etiologically implicated in Alzheimer's disease (AD). The impetus for this study is based on the need for the application of strong epidemiologic methods to an hypothesis that has been one of the most enduring and controversial hypotheses in AD. The foundation for the proposed study has been laid by data from a previous case-control study conducted by the proposed Principal Investigator, which evaluated a variety of risk factors for AD. In that study, the use of aluminum- containing antiperspirants and antacids was found to confer an increased relative risk of developing AD. We now have the opportunity to investigate these as well as a number of other sources of aluminum exposure in a unique group of population-based patients with short disease duration. Data on 180 clinically diagnosed cases enrolled in the Alzheimer's Disease Patient Registry at Group Health Cooperative in Seattle, WA. Will be compared to an equal number of controls selected from the enrollment rolls of this health maintenance organization. Controls are screened to rule out cognitive impairment based on an ongoing study to examine platelet membrane fluidity differences between cases and controls. Cases and controls will frequency matched by age, sex, and year of onset of the disease in the case and selection of subjects will be restricted to those who have been married for at least ten years before they were first identified by the research team. In- person interviews with the case and control spouse and with the controls themselves will be performed using a photograph album to assist in recall of prior use of antiperspirants, antacids and analgesics, and using typed cards with brand names for other items. The major exposure categories to be studied are (a) antiperspirants, antacids and analgesics, (b) commonly used household items such as aluminum cookware, food packaging and selected aluminum-containing foods, and (c) a detailed occupational history which will involve ranking aluminum exposure levels on each job. Additional data to be used to study confounding and effect-modification will be available on age, family history of dementia, history of prior head injuries, thyroid disease, and other variables being collected as part of the platelet membrane fluidity study. Data will be analyzed using logistic regression. Because this study yield a more accurate estimate of the true association between aluminum and AD than currently exists in the literature.
