Does an interaction between smoking and drinking influence the risk of Alzheimer's disease? Results from three Canadian data sets

Investigation of the relationship of smoking and drinking to Alzheimer's di sease (AD) may advance research on the cause of AD and provide a basis for treatment. Pharmacological mechanisms for an involvement of smoking and dri nking are plausible but epidemiologic reports are inconsistent. Evidence of behavioural and physiological interactions suggests that tobacco and alcoh ol use may not only individually affect AD, but may also modify each ether' s effects. A modelling strategy was developed to examine the interaction be tween smoking and drinking on the risk of AD. Three Canadian data sets were analysed: the University of Western Ontario Dementia Study (UWODS) (n = 36 3); the Canadian Study of Health and Aging (CSHA) (n = 516), and the databa se from the Clinic for Alzheimer Disease and Related Disorders at the Vanco uver Hospital and Health Sciences Centre, University of British Columbia si te (UBC) (n = 843). Multiple logistic regression models were adjusted for t he potential confounders age, age squared, sex, education, family history o f dementia, head injury and hypertension. Analysis of the CSHA provided evi dence consistent with the hypothesis that smoking and drinking influence ea ch other's effects on AD, with smoking reducing the risk of AD among drinke rs. A similar interaction was marginally significant (p = 0.052) in the UWO DS data set, but not significant in the UBC data. Extension of these analys es, particularly in longitudinal studies and within genetic risk groups, is needed to determine whether this interaction can be replicated. If so, res earch on the biological interactions of nicotine and alcohol may provide a basis for the development of therapeutic interventions as well as providing clues to the cause of this disorder. Copyright (C) 2000 John Wiley & Sons, Ltd.