PURPOSE: The South Korean (SK) government monopolizes the tobacco industry and is accused of pushing smoking on captive military personnel. However, estimating the association between military service and smoking is difficult, since military service is required for all SK men and the few civilian waivers are usually based on smoking determinants, e.g., social status. MATERIALS AND METHODS: Using a quasi-experimental design we validly estimate the association between military service and smoking. Military service was assigned by immigration patterns to the United States, instead of an experimenter, by comparing Korean Americans who happened to immigrate before or after the age(s) of mandated service. Smoking promotion in the military was also described among SK veterans, to identify the probable mechanisms for veterans' smoking tendencies. RESULTS: Veterans were 15% [95% confidence interval (CI), 4 to 27] more likely to ever-puff and 10% (95% CI, 0 to 23) more likely to ever-smoke cigarettes, compared to a similar group of civilians. Among veterans, 92% (95% CI, 89 to 95) recalled cigarettes were free, 30% (95% CI, 25 to 35) recalled smokers were given more work breaks and 38% (95% CI, 32 to 43) felt explicit "social pressure" to smoke. Free cigarettes was the strongest mechanism for veterans' smoking tendencies, e.g., veterans recalling free cigarette distribution were 16% (95% CI, 1 to 37) more likely to ever-smoke than veterans not recalling. CONCLUSION: These patterns suggest military service is strongly associated with smoking, and differences between veterans and civilians smoking may carry over long after military service. Given military service remains entirely in government purview, actively changing military smoking policies may prove most efficacious. This highlights the importance of recent bans on military cigarette distribution, but policies eliminating other smoking encouragements described by veterans are necessary and could effectively reduce the smoking prevalence by as much as 10% in SK.
Adolescent ; Adult ; Aged ; Humans ; Male ; Middle Aged ; Military Personnel/*statistics & numerical data ; Republic of Korea/epidemiology ; Smoking/*epidemiology ; Young Adult