Despite a widespread belief that glucocorticoid therapy is associated with positive energy balance and excess weight gain there is a dearth of quantitative evidence about its effects and the underlying mechanisms of any effects. The primary aim of the present study was to quantify the effect of dexamethasone and prednisone treatment on energy intake in children treated for childhood acute lymphoblastic leukemia. A secondary aim was to test for differences in excess weight gain between patients treated using the 2 glucocorticoids. We measured energy intake in 26 patients (mean +/- SD age, 6.3 +/- 2.3 yr) during a 5-d period "on" steroids and again in the week before steroid treatment. Changes in body mass index from diagnosis to 1 and 2 yr postdiagnosis were expressed as SD scores. Steroid treatment was associated with a significant increase in energy intake of approximately 20% (mean paired difference, 1.7 MJ/d; SD, 2.8; 95% confidence interval, 0.7-2.8 MJ/d), with no significant difference between the 2 steroids. The mean change in body mass index SD score was +0.38 (SD, 1.10; P < 0.05) to 1 yr and +0.68 (SD, 1.38; P < 0.05) to 2 yr, with no significant difference between the 2 groups of patients. Glucocorticoid treatment in childhood acute lymphoblastic leukemia increases energy intake markedly, and this effect contributes to the excess weight gain and obesity characteristic of patients being treated for acute lymphoblastic leukemia.