Dietary fat as an exposure in epidemiological studies. Observations in the Malmö Diet and Cancer cohort

Sammanfattning: The aim was to explore the inherent meaning of total fat and to examine the associations between relative fat intake and postmenopausal breast cancer. Furthermore, the impacts of the alterations in dietary assessment routines are discussed. This thesis uses data from the Malmö Diet and Cancer study, a prospective cohort study. Information on dietary habits, lifestyle, and socioeconomic factors was collected between 1991 and 1996. The associations between fat intake, lifestyle factors, sociodemographic factors and intakes of food groups and nutrients were examined in a sup-sample of 10 295 individuals. The associations between intakes of relative fat, alcohol, plant foods and fiber and breast cancer risk were examined in a sub-sample of 11 726 postmenopausal women. High fat intake was consistently associated with smoking, low leisure time physical activity and living alone. There were differences in intakes of almost all food groups and nutrients across quintiles of relative fat. We found an increased breast cancer risk associated with high alcohol intake. Diets characterised by high fiber and low fat intakes were associated with decreased risk. Conclusion: Results suggest that many food groups, plant foods especially; nutrients and other lifestyle factors may confound associations between relative fat intakes and disease. High alcohol intakes increased the risk of postmenopausal breast cancer. Dietary habits characterised by high fiber and low fat intakes are associated with breast cancer protection. In this thesis, the residual energy-adjustment model was found to have advantages compared to the other models.