The empirical evidence concerning the quantitative relations between energy intake and expenditure on the one hand and body mass index (BMI) on the other is inconclusive. We have used a large database of 14,281 individuals, for whom habitual dietary intake and expenditure have been ascertained with adequate methodology, to examine the mutually adjusted associations of these variables with BMI. Study subjects were adult participants in the Greek component of the European Prospective Investigation into Cancer and Nutrition (EPIC) study. After adjustment for age and gender and exclusion of dieting individuals and energy underreporters, an increment of about 500 kcal intake was found to correspond to an increment of about 0.33 kg/m2 of BMI, whereas an increment of about 5 MET-hours of energy expenditure was associated with a decrease of about 0.18 kg/m2 of BMI, where MET is the ratio of the metabolic rate associated with a given activity to the resting metabolic rate. Our results indicate that increasing physical activity is about half as effective as decreasing energy intake in reducing BMI.