Functional translation of exercise responses from exercise testing to exercise training: The test of a model

Exercise prescription based on exercise test results is complicated by the need to down-regulate the absolute training intensity to account for cardiovascular drift in order to achieve a desired internal training load. We tested a recently developed generalized model to perform this downregulation using metabolic equivalents (METs) during exercise testing and training. A total of 20 healthy volunteers performed an exercise test to define the METs at 60, 70, and 80% of the heart rate (HR) reserve and then performed randomly ordered 30 min training bouts at absolute intensities predicted by the model to achieve these levels of training intensity. The training HR at 60 and 70% HR reserve, but not 80%, was significantly less than predicted from the exercise test, although the differences were small. None of the ratings of perceived exertion (RPE) values during training were significantly different than predicted. There was a strong overall correlation between predicted and observed HR (r = 0.88) and RPE (r = 0.52), with 92% of HR values within ±10 bpm and 74% of RPE values within ±1 au. We conclude that the generalized functional translation model is generally adequate to allow the generation of early absolute training loads that lead to desired internal training loads.