International comparisons of nutrition and mortality from cancers of the oesophagus, stomach and pancreas.

The average per capita consumption of certain foods in 29 countries during 1964-66 was related to the average mortality rates for cancers of the oesophagus, stomach and pancreas in these same countries for the period 1978-79. No significant correlation was found between consumption of nutrients from different food groups and mortality rates for oesophageal cancer, with the exception of meat intake in females. Consumption of eggs, milk and particularly meat was inversely related to mortality from stomach cancer in males and females. Caloric and fat intakes from animal sources were also significantly and negatively related to mortality from stomach cancer in both sexes, while caloric intake from vegetable sources was positively associated, particularly in females. The absence of correlation between vegetable fat intake and mortality from stomach cancer strongly suggests carbohydrates as a major caloric source related to stomach cancer. Mortality from pancreatic cancer was strongly and positively related to consumption of eggs, milk and meat in both sexes. Caloric and fat intakes from animal sources had a strong positive correlation with mortality from pancreatic cancer in males and females, while caloric intake from vegetable sources showed a strong negative correlation in both sexes. The weak and not significant correlation of vegetable fat with mortality from pancreatic cancer mortality strongly suggests carbohydrates as a major caloric source related to pancreatic cancer. No significant correlation was found between age-adjusted mortality rates from stomach or pancreatic cancer or a given per capita caloric intake from animal or vegetable source. For both cancers, age-adjusted mortality rates were about twice those of females.