Individualising training intensity to reduce inter-individual variability in training response in trained cyclists

C O' Grady; J Hopker

Authors

C O' Grady Endurance Research Group, School of Sport and Exercise Sciences, University of Kent, UK
J Hopker Endurance Research Group, School of Sport and Exercise Sciences, University of Kent, UK

Keywords:

training, individualisation, high-intensity, cycling, individual variability.

Abstract

Background: Training to improve endurance performance commonly results in large inter-individual variability (IIV) in response (Bouchard et al. [1998]. Medicine and Science in Sports and Exercise, 30(2), 252–258; Mann et al. [2014]. Sports Medicine, 44, 1113–1124). A novel perspective to this issue centers on the differences in physiological response at set percentages of maximal performances; commonly used to prescribe training (Coyle et al. [1988]. Journal of Applied Physiology, 64(6), 2622–2630). By establishing individual profiles of performance using a Power Law (PL), training intensity could be prescribed on an individualised basis (García-Manso et al. [2012]. Journal of Theoretical Biology, 300, 324–329).

Purpose: This investigation sought to determine whether using a PL could reduce IIV in V̇O_2max response to training compared to using a standardised method.

Methods: Two groups of male cyclists completed 12 high intensity training (HIIT) sessions over 4 weeks. Training intensity was prescribed using PL models in the individualised group (IG; n=5, V̇O_2max = 57.50 ± 9.02 mL.kg.min^-1) and set percentages of V̇O_2max in the standardized group (SG; n=5, V̇O_2max = 62.17 ± 4.45 mL.kg.min^-1). A V̇O_2max test and performance time trial were completed pre- and post-training. PL’s were established using maximal efforts of 12, 7, and 3 minutes (Galbraith et al. [2014]. Journal of Sports Physiology and Performance, 9(6), 931–935). Training sessions consisted of 3 sets of 10 repetitions of 30 seconds work and 30 seconds recovery, with 5 minutes active recovery between sets. Statistical analyses were conducted using IBM SPSS Statistics 22, with between- and within-group comparisons completed using independent and paired samples t-tests, respectively. Variability was analysed using log-transformed coefficients of variation and Bland-Altman plots.

Results: V̇O_2max was shown to have significantly increased in IG from 57.50 ± 9.02 mL.kg.min^-1 to 59.36 mL.kg.min^-1 following 4 weeks of HIIT training prescribed using a PL (P < 0.05). V̇O_2max did not significantly improve in SG (P > 0.05; Figure 1). Intra-class correlation coefficients (ICC) showed that variability in V̇O_2max response in both IG and SG was low, but significantly stronger correlations were observed in IG (P < 0.001) than in SG (P < 0.05). Individual V̇O_2max response profiles (Figure 2) indicate wider variation in response in SG, with two participants showing reduced V̇O_2max, and a more consistent positive response in IG. Bland-Altman plots identify variance in V̇O_2max response of + 4.39 ml.kg.min^-1 to - 0.69 ml.kg.min^-1in IG and from + 8.86 ml.kg.min^-1 to – 6.23 ml.kg.min^-1 in SG (Figure 3).

Conclusion: The results of this study suggest that individualised HIIT training prescribed using a PL can reduce the IIV in V̇O_2max response to training when compared to a standardised approach. This indicated that prescribing training using a PL model can result in consistent and predictable responses, useful for research, clinical, and applied purposes.