Background: The relationship between diet and cancer of the pancreas was assessed in a ecological study by calculating the intakes of foods and nutrients in the different Spanish provinces, during the period 1964-65, and relating these to the provincial pattern of death from pancreatic cancer 20 years after (1984-86).
Methods: The geographical pattern of mortality was evaluated by calculating the Standardized Mortality Ratios (SMR) for each province. Intake of foods was obtained from the National Institute of Statistics. Nutrients intakes were calculated from these consumption figures, by reference to tables of food composition. Simple and multiple linear regression studies with these variables were carried out.
Results: Regression coefficients obtained with foods and nutrients in the univariate analyses were in general low. In the separate multivariate models, consumption of milk and cheese were constantly positively correlated with pancreatic cancer mortality rates in males. For females, consumption of eggs was positively associated, and fruit consumption was negatively correlated. Analyses of nutrient intake adjusted for total energy showed that proteins were the most strongly correlated of the variables considered in both sexes. Animal fat, cholesterol and saturated fat were statistically significant positively correlated with pancreatic cancer mortality.
Conclusions: The results obtained are consistent with the hypothesis that diet can play a role in the etiology of the cancer of the pancreas. Further studies are needed to obtain empirical evidence for (or against) the associations found.