BACKGROUND Helicobacter pylori is a major cause of various gastroduodenal diseases. Some risk factors related to H. pylori infection have been reported; however, studies on the relationship between H. pylori infection and smoking or drinking habits have given conflicting results. In the present study, these relationships were investigated by collecting sera and information from 8837 subjects. METHODS Serum H. pylori immunoglobulin G antibody was measured by an enzyme-linked immunoassay. In addition to sex and age, information on smoking and drinking habits was collected by questionnaire. Age- and sex-adjusted odds ratios (95% confidence interval) of smoking and alcohol consumption were calculated for H. pylori seropositivity using logistic regression models. RESULTS Current smokers had a 0.82 (0.74-0.91)-fold greater risk of H. pylori seropositivity than those who had never smoked. Current cigarette consumption showed a dose-dependently negative association with H. pylori seropositivity, and the association between smoking and H. pylori infection was strong in younger subjects. Current drinkers had a 0.88 (0.79-0.98)-fold greater risk of H. pylori seropositivity than those who had never drunk alcohol. The volume of alcohol consumed showed a negative association with H. pylori seropositivity. CONCLUSIONS In the current study, smoking was negatively associated with H. pylori infection. The risk of H. pylori seropositivity decreased linearly with cigarette consumption per day. Increased gastric acidity in the stomach through smoking may be a cause of the dose-dependently negative association between H. pylori and smoking. Drinking was negatively and dose-dependently associated with H. pylori positivity, although the effect of drinking was weaker than that of smoking.