This study evaluated the dietary pattern of foods and nutrients according to levels of vigorous leisure time physical activity (PA) assessed at recruitment within the Spanish cohort of the European Prospective Investigation on Cancer (EPIC) study (37,287 healthy volunteers with complete information). We used a validated PA questionnaire (PAQ) to measure the weekly frequency and duration of different kinds of sport activities. For dietary assessment, we used a validated diet history questionnaire that included all items consumed with a frequency of at least twice a month. We tested differences in food and nutrient intake according to PA duration by means of both an analysis of variance and an analysis of covariance adjusted for confounding factors. Linear increases or decreases in food and nutrient intake across PA levels were tested by means of a regression analysis. Only 11% of men and 6% of women performed at least 3 hours/week of intense PA, which is similar to current recommendations. Overall, main nutrient and total energy intakes were similar across different PA levels (<2% change in total energy intake between extreme PA categories). However, the intake of some foods and vitamins did significantly (p ≤ .05) increase as PA increased. The average gender-weighted percentage change in the intake of food and vitamins increased when moving from the lowest levels of PA to the highest. There was an increase in the intake of the following: 15.9% in vegetables, 6.7% in fruit, 9% in fish, 5.6% in dairy products, 10% in vitamin C, 5.9% in vitamin E, 7.2% in retinol, 19.7% in total carotene, 40.1% in α-carotene, 20.4% in ß-carotene, 11.2% in licopene, and 26.1% in lutein. BMI, which was above average for the cohort (mean ± SD: 28.4 ± 4.2), decreased steadily when PA increased. To sum up, in this large Spanish cohort, the differences in dietary intake relative to levels of PA were not found either in the amount of total energy consumed nor in the number of main macronutrients but rather in the intake of certain foods which, while having very little or moderate caloric content, are very rich in highly bioactive elements such as vitamins and provitamins.
M.J. Tormo, C. Navarro, M.D. Chirlaque, and X. Barber are with the Consejería de Sanidad y Consumo, Ronda de Levante, 11 E-30008, Murcia, Spain. S. Argilaga, A. Agudo, and C.A. González are with the Instituto Catalán de Oncología, Barcelona, Spain. P. Amiano, J.M. Beguiristain, and M. Dorronsoro are with the Dirección de Salud de Gipúzkoa, San Sebastian, Spain. A. Barricarte is with the Departamento de Salud de Navarra, Pamplona, Spain. C. Martínez and M. Rodríguez are with the Escuela Andaluza de Salud Pública, Granada, Spain. J.R. Quirós is with the Consejería de Sanidad y Servicios Sociales de Asturias, Oviedo, Spain.