Several prediction equations developed to convert body movement measured by accelerometry into energy expenditure have been published. The aim of this study was to examine the degree of agreement between three different prediction equations, when applied to data on physical activity in a large sample of children. We examined 1321 children (663 boys, 658 girls; mean age 9.6+/-0.4 years) from four different countries. Physical activity was measured by the MTI accelerometer. One equation, derived from doubly labeled water (DLW) measurements, was compared with one treadmill-based (TM) and one room calorimeter-based (CAL) equation (mixture of activities). Predicted physical activity energy expenditure (PAEE) was the main outcome variable. In comparison with DLW-predicted PAEE, both laboratory-derived equations significantly (P<0.001) overestimated PAEE by 17% and 83%, respectively, when based on a 24-h prediction, while the TM equation significantly (P<0.001) underestimated PAEE by 46%, when based on awake time only. In contrast, the CAL equation agreed better with the DLW equation under the awake time assumption. Predicted PAEE differ substantially between equations, depending on time-frame assumptions, and interpretations of average levels of PAEE in children from available equations should be made with caution. Further development of equations applicable to free-living scenarios is needed.