The energy cost of horizontal walking and running in adolescents.Med. Sci. Sports Exerc.,Vol. 31, No. 2, pp. 311-322, 1999.Purpose:This study developed and cross-validated generalized equations for predicting &OV0312;O2(mL·kg−1·min−1) and caloric expenditure (kcal·kg−1·min−1) during horizontal walking and running in adolescents.Methods:Subjects were 47 male and 35 female adolescent volunteers, ages 12-18. Each subject underwent a submaximal treadmill exercise test to determine &OV0312;O2at randomly selected walking and jogging speeds (67-215 m·min−1). Caloric expenditure was estimated from &OV0312;O2and RER. Multiple regression was used to develop prediction equations for estimating &OV0312;O2and caloric expenditure from a derivation sample of 77 random observations, both walking and running.Results:The group relationship between running speed and energy cost in the derivation sample was linear, whereas the relationship between walking speed and energy cost was quadratic. Gender, age, and height each failed to account for significant additional variation in energy cost after speed and mode were considered. Skinfolds accounted for a small yet significant amount of additional variation in energy cost. The derived equations were cross-validated on a sample of 76 separate random observations. The cross-validation statistics are: for &OV0312;O2, R = 0.95, error = 3.58 mL·kg−1·min−1, and for caloric expenditure, R = 0.94, error = 0.019 kcal·kg−1·min−1. Most selected adult equations consistently underestimated both &OV0312;O2and caloric expenditure in the cross-validation sample.Conclusions:These results suggest that in adolescents, within the range of speeds tested, the relationship between speed of movement and energy cost for running is linear, but for walking is curvilinear. Also, adult models for estimating &OV0312;O2or caloric expenditure do not account for the higher relative energy cost of walking and running in adolescents.