Objective: To assess the mortality risks from injuries for smokers and ex-smokers and to quantify the mortality burden of smoking from injury in Taiwan. Methods: Smokers' (and ex-smokers') mortality risks from injuries were compared with that of nonsmokers in a merged cohort from Taiwan. A total of 64 319 male subjects were followed up for 12 - 18 years. Relative risks (RR) (adjusted for age and alcohol use) and 95% confidence intervals (CI) for cause specific injury deaths were calculated using the Cox proportional hazard model. Relative risks of injury mortality were also calculated to assess the presence of dose - response relations with daily smoking quantity. Results: Alcohol use adjusted relative mortality risks for all injuries (RR 1.69, 95% CI 1.39 to 2.05) including those from motor vehicle accidents (RR 1.88, 95% CI 1.44 to 2.45) and non- motor vehicle accidents (RR 1.48, 95% CI 1.11 to 1.99) were significantly higher for smokers than non- smokers. Mortality was also increased for most subtypes of non- motor vehicle injuries including falls, fires, and job related injuries. Furthermore, these increases were dose dependent, with the heaviest smokers having the highest risk and the lightest smokers the lowest risk, and ex-smokers, no increase. In 2001, over one fifth (23%) of all male injury deaths in Taiwan was associated with smoking. Conclusion: This study demonstrated the significant association between fatal injuries and smoking. This relation adds further weight to smoking cessation campaigns.
