Abstract
BACKGROUND: Although smoking and alcohol drinking are established risk factors of esophageal cancer, their public health impact is unclear. Furthermore, the effect of green tea is controversial. METHODS: The present study was based on a pooled analysis of two prospective cohort studies. A self-administered questionnaire about health habits was distributed to 9,008 men in Cohort 1 and 17,715 men in Cohort 2, aged 40 years or older, with no previous history of cancer. We identified 38 and 40 patient cases with esophageal cancer among the subjects in Cohort 1 (9.0 years of follow-up) and Cohort 2 (7.6 years of follow-up), respectively. Cox proportional hazards regression was used to estimate hazard ratios (HRs) of the risk of esophageal cancer incidence. RESULTS: Cigarette smoking, alcohol drinking and green tea consumption were significantly associated with an increased risk of esophageal cancer. Compared with men who had never smoked, never drunk alcohol or green tea, the pooled multivariate HRs (95% confidence intervals) were 5.09 (1.80-14.40) (p for trend <0.0001), 2.73 (1.55-4.81) (p for trend=0.0002), or 1.67 (0.89-3.16) (P for trend=0.04) for men who were currently smoking > or =20 cigarettes/day, drinking alcohol daily, or drinking > or =5 cups green tea/day, respectively. The population attributable fractions of esophageal cancer incidence that was attributable to smoking, alcohol drinking and green tea consumption were 72.0%, 48.6%, and 22.1%, respectively. CONCLUSIONS: Among the variables studied, smoking has the largest public health impact on esophageal cancer incidence in Japanese men, followed by alcohol drinking and green tea drinking.