In boys with DMD, muscle weakness progresses in a proximal to distal pattern, leading to compensatory gait strategies, including hyperlordosis and equinus, that increase energy cost and accelerate the loss of walking capacity. The purpose of this study was to determine the changes in the energy cost of walking that occur with disease progression and to determine the optimal normalization scheme for the longitudinal assessment of the energy cost of walking in boys with DMD. Energy cost was assessed with the COSMED K4b2. Three normalization schemes were examined: gross energy cost (EC), net non-dimensional oxygen cost (NNcost) and speed-matched control energy cost (SMC-EC). Nonlinear mixed modeling procedures determined the rate of change with age. Linear regression was used to asses the relationship between each normalization scheme and age and body height. 74 boys with DMD were assessed for the energy cost of walking. Velocity decreased at a significant rate (-.00245/month, p = .03) across time; (Fig. 2), while gross EC (.003248/month, p = 0.0026), NNcost (.006155/month, p < 0.0001) and SMC-EC (.001690/month, p = 0.03) all increased significantly. Age and height were significantly associated with NNcost and SMC-EC. The sensitivity of NNcost and SMC-EC to age over time were similar, while SMC-EC was less sensitive to changes in height over time than NNcost. In contrast to able-bodied peers, boys with DMD decrease their velocity while all walking energy cost measures increased over time. Both SMC-EC and NNcost proved appropriate normalization schemes for boys with DMD. Compared to gross EC, both NNcost and SMC-EC were less sensitive to changes in age over time, while SMC-EC was less sensitive to changes in height than NNcost. Therefore, both NNCost and SMC-EC are suggested normalization schemes for the longitudinal assessment of energy cost in boys with DMD.