Conventional methods to prescribe exercise intensity are ineffective for exhaustive interval training.

Purpose. To compare methods of relative intensity prescription for their ability to normalise performance (i.e. time to exhaustion), physiological, and perceptual responses to high-intensity interval training (HIIT) between individuals.

Methods. Sixteen male and two female cyclists (age: 38 ± 11 years, height: 177 ± 7 cm, body mass: 71.6 ± 7.9 kg, maximal oxygen uptake (V̇ O2max): 54.3 ± 8.9 ml·kg-1 ·min-1 ) initially undertook an incremental test to exhaustion, a 3-min all-out test, and a 20-min time-trial to determine prescription benchmarks. Then, four HIIT sessions (4-min on, 2-min off) were each performed to exhaustion at: the work rate associated with the gas exchange threshold (ẆGET) plus 70% of the difference between ẆGET and the work rate associated with V̇ O2max;85% of the maximal work rate of the incremental test (85%Ẇmax); 120% of the mean work rate of the 20-min time-trial (120%TT); and the work rate predicted to expend, in 4 min, 80% of the work capacity above critical

power. Acute HIIT responses were modelled with participant as a random effect to provide estimates of inter�individual variability.

Results. For all dependent variables, the magnitude of inter-individual variability was high, and confidence intervals overlapped substantially, indicating that the relative intensity normalisation methods were similarly poor. Inter-individual coefficients of variation for time to exhaustion varied from 44.2% (85%Ẇmax) to 59.1% (120%TT), making it difficult to predict acute HIIT responses for an individual.

Conclusion. The present study suggests that the methods of intensity prescription investigated do not normalise

acute responses to HIIT between individuals.