RESUMO
OBJECTIVE: Alcohol drinking and tobacco smoking have impacts on lifestyle-related diseases, but their association with dementia remains a debated topic. This study aimed to examine longitudinal associations between alcohol consumption, smoking, and dementia risk in middle-aged and older Japanese people. METHODS: This study used a cohort design with an 8-year follow-up. Participants were community-dwelling Japanese people (N = 13,802) aged 40-74 years. The baseline survey, including a self-administered questionnaire, was conducted in 2011-2013. Predictors were alcohol consumption and tobacco smoking. The outcome was incident dementia obtained from a long-term care insurance database. Covariates were demographics, lifestyle factors, body mass index, general health status, and history of stroke, diabetes, and depression. RESULTS: Participant mean age was 59.0 years. The 1-149, 150-299, and 300-449 g ethanol/week groups had significantly lower adjusted hazard ratios (HRs) (0.62, 0.59, and 0.47, respectively) compared with the reference group, with no significant linear association. HRs increased toward 1 when past-drinkers and those with poor health status and a disease history were excluded (0.80, 0.66, and 0.82, respectively). Higher smoking levels were dose-dependently associated with a higher HR (adjusted P for trend = 0.0105), with the ≥20 cigarettes/day group having a significantly higher adjusted HR (1.80). Heavy drinkers (≥449 g ethanol/week) with smoking habits, but not those without smoking habits, had higher dementia risk (P for interaction = 0.0046). CONCLUSION: Light-to-moderate alcohol consumption is associated with decreased dementia risk, and smoking is dose-dependently associated with increased dementia risk, with an interaction between high alcohol consumption and smoking on dementia risk.