Individualised Methods of Prescribing Exercise in Cycling

Training is a complex, multi-factorial process, which involves the manipulation of the duration, frequency and intensity of exercise. When quantifying the physiological and performance responses to training a large inter-individual variability in training responses is frequently observed. To date, the majority of research has examined the relationship between genetics and trainability. Another hypothesis, which has not been fully explored, is that the variability is also due to an inappropriate standardisation of exercise intensity or duration. This thesis, therefore, presents a series of studies that investigate the effects of individualised methods of prescribing exercise intensity and duration on performance and physiological responses in cycling.

Study 1 compared time-to-exhaustion (TTE) to time-trial (TT) performances when the duration of the trials were matched and participants were blinded to feedback. A higher mean power output was found for TTE compared to TT at 80% (294 ± 44 W vs. 282 ± 43 W respectively, P<0.05), but not at 100% (353 ± 62 W vs. 359 ± 74 W) and 105% (373 ± 63 W vs. 374 ± 61 W) of maximum aerobic power (MAP). Critical power (CP) calculated from the TTE trials was also higher, whereas, anaerobic work capacity (W?) was lower (P<0.05). The findings favour TTE over TT performances for a higher mean power output and calculated CP.

Study 2 compared the effects of three training intensities: moderate intensity (MOD), high intensity (HIT) and a combination of the two (MIX) when the duration of exercise was individualised. Participants were randomly assigned to one training group and trained 4 times per week for 4-weeks. Training duration was individualised to each participant’s maximum performance. All training groups increased maximal oxygen uptake (V?O2max), MAP, TTE and gross efficiency (GE) after training (P<0.05), but no differences were observed between groups (P>0.05). Therefore, when the duration of training is individualised, similar improvements in performance and physiological responses are found, despite differences in exercise intensity.

The CP and power law models propose power-duration relationships that describe maximum endurance capacity. Study 3 compared the predictive ability of these two models for TTE performances. It was hypothesised that the CP and power law models would reliably predict actual TTE for intensities between 80-110% MAP, but a power

law model would better predict TTE for intensities outside of this range. No significant differences for parameter estimates were found between models (CP and power law) and actual TTE for intensities ranging from 80-110% MAP. Outside of this range however, the CP model over predicted actual performance at 60% and 150% MAP (P<0.05), while there was no significant difference between the power law model and actual performance at these intensities (P>0.05). Both models were different from actual performance at 200% MAP (P<0.05). Therefore, a power law model can accurately predict cycling TTE for intensities ranging from 60-150% MAP.

Study 4 tested the hypothesis that the inter-individual variability for TTE performances is due to the methods used to standardise exercise intensity. A %V?O2max prescription was compared with an alternative based on an individual power-duration relationship (using a power law model). A power law model predicted the intensity for TTE lasting exactly 20-min and 3-min. A corresponding intensity for TTE as a %V?O2max was 88% and 109%. On two separate occasions participants completed two TTE trials using the power law and %V?O2max prescriptions, with 30-min rest between trials. There was a significant reduction in the inter-individual variability for TTE when exercise was prescribed using a 20-min power law versus 88% V?O2max prescription method (coefficient of variation = 29.7 vs. 59.9% respectively; P<0.05). However, there was no significant difference in the inter-individual variability for TTE using a 3-min power law versus 109% V?O2max prescription method (P>0.05).

Two main conclusions can be drawn from this thesis. Firstly, a power law model can accurately predict and describe cycling endurance performance across a wide range of intensities. Secondly, prescribing exercise intensity using a power law model reduces the variability in TTE by 50% when compared to a %V?O2max prescription method. Therefore, the methods used to standardise exercise intensity appear to be related to the variability in TTE performances. Future research should examine whether training prescribed using a power law model reduces the variability in subsequent training responses.