During the last 20 years, numerous case-control or epidemiological studies have examined the environmental and genetic risk factors of Alzheimer's disease. A review of available evidence indicates that smoking and chronic use of anti-inflammatory drugs might be associated with a decreased risk of Alzheimer's disease. However, this hypothesis is supported by data collected in case-control studies or in prevalence studies and must be confirmed in prospective cohort studies. About ten per cent of Alzheimer's disease cases are characterized by early onset and familial aggregation. Most of these cases have been recently linked to the presemlin 1 gene which is located on chromosome 14; in other families, the disease is associated is a mutation of the beta amyloid precursor gene on chromosome 21. The risk of late-onset sporadic cases depends upon the polymorphism of the apolipoprotein E gene; allele E4 carriers have a 4-time increased risk.