Objective: To study risk factors for Alzheimer's disease (AD) based on
data from the Canadian Study of Health and Aging. Design: Population-
based case-control study. Setting: Communities and institutions in 10
Canadian provinces. Participants: Two hundred fifty-eight cases clinic
ally diagnosed with probable AD, with onset of symptoms within 3 years
of diagnosis, and 535 controls, frequency matched on age group, study
center, and residence in community or institution, clinically confirm
ed to be cognitively normal. Main outcome measure: Odds ratios (ORs) w
ere calculated using unconditional logistic regression for previously
hypothesized and potential risk factors for AD. Results: The OR for fa
mily history of dementia was significantly elevated (2.62; 95% confide
nce interval [CI], 1.53 to 4.51) and increased with the number of rela
tives with dementia. Those with less education were at higher risk of
AD, with an OR of 4.00 (95% CI, 2.49 to 6.43) for those with 0 to 6 ye
ars, in comparison with those with 10 or more years. Head injury achie
ved borderline significance. A history of arthritis resulted in a low
risk of AD (OR = 0.54; 95% CI, 0.36 to 0.81), as did a history of use
of nonsteroidal anti-inflammatory drugs, Initial analyses showed an in
creased risk of AD for occupational exposure to glues as well as to pe
sticides and fertilizers; the increased risk was greater in those with
less education. Conclusion: This study confirmed a number of previous
ly reported risk factors for AD, but provided little support for other
s. A new finding was an increased risk for those with occupational exp
osure to glues as well as pesticides and fertilizers, but this needs f
urther study.