Coffee and pancreatic cancer: an analysis of international mortality data.

This study assessed the relationship of per capita coffee imports and consumption, total dietary fat, saturated fat, cholesterol, tobacco, cigarettes, and national income for 1957-1965 to age-adjusted pancreatic cancer death rates of men and women from 22 countries in 1971-1974. With simple correlation analysis, coffee, total dietary fat, saturated fat, and national income were each significantly correlated with both male and female pancreatic cancer mortality. Bivariate partial correlation coefficients of coffee with pancreatic cancer mortality were significant (one-tailed) in 11 of 12 analyses and borderline significant in two-way analyses of variance (ANOVA) (two-tailed) controlling for each of the other variables. Saturated fat and pancreatic cancer were also significantly related in univariate analyses, and in 11 of 12 bivariate partial correlation analyses; in ANOVA, significance was borderline in 10 of 12 analyses. Total fat and pancreatic mortality were also significantly associated in most of the univariate and bivariate correlation analyses, but not in the two-way analyses of variance. The findings of this study are consistent with the hypothesis that coffee and dietary lipid are involved in the etiology of pancreatic cancer.