The association between dementia and smoking or alcohol use has been examined in several epidemiological studies. In many case-control studies, a decreased risk of dementia or Alzheimer's disease (AD) was observed among smokers. However, when this association was analysed in prospective studies, an increased risk of AD was observed. In addition, in the PAQUID study, we showed that the decreased risk disappeared after adjustment for educational level and occupation. These factors are strong confounders in the association between dementia and tobacco use. We also showed that moderate consumption of wine was associated with a lower risk of developing AD. This result remains unchanged after adjustment for many potential confounders. The association between moderate alcohol consumption and risk of developing a dementia or AD was recently confirmed by prospective studies. In some studies, wine consumption was more specifically associated with a decreased risk, whereas beer or spirit consumption was not associated. These results suggest that tobacco consumption is not associated with a lower risk of dementia and that moderate alcohol intake does not increase the risk of developing dementia.