The effect of age, sex, and education on food consumption of a middle-agedEnglish cohort - EPIC in East Anglia

Background Different dietary patterns are associated with differing risks o f chronic disease. Yet independent relationships between diet and demograph ic variables, such as age, sex, and education, are poorly described. Methods. The first 1968 subjects enrolled to the European Prospective Inves tigation of Cancer (EPIC) cohort from general practices in East Anglia, UH, provided food frequency and demographic data. Results. Men ate meat, eggs, milk, and sugary foods more frequently, but fr uit and vegetables less frequently than women. Cider subjects ate red meats and saturated bread spreads more frequently but consumed less poultry and drank less coffee than younger subjects. Better educated subjects ate less meat, more salads, and fewer cakes and sweet foods than those less educated . Five clusters representing different dietary patterns were readily identi fied. These were (a) younger well-educated, probably containing many vegeta rians; (b) "low calorie," two-thirds female; (c) high alcohol, nuts, meat, largely male; (d) preferring fruits, vegetables, unsaturated fats, poultry; and fish, 71% female; (e) preferring meat, potatoes, sweet foods, saturate d fats, less well-educated older men. Conclusions. The reported consumption of many foods varies by age, gender, and education. A pattern of eating that is generally considered less health ful was particularly seen in older men, placing them at increased risk of c hronic disease, (C) 2000 American Health Foundation and Academic Press.