Change in VO2max and time trial performance in response to high-intensity interval training prescribed using ventilatory threshold

Abstract

A greater volume of HIIT may be needed to maximize the training response for all individuals.

Moving beyond threshold-based dichotomous classification to improve the accuracy in classifying non-responders

“…; Montero and Lundby ; Astorino TA et al. ), threshold‐based dichotomous classification approaches misclassify individuals who have a high probability of a positive response as non‐responders, and 2) using response CIs with zero‐ and/or SWC‐based thresholds provides more information than dichotomous classification approaches regarding an individual's response. Application of these statistical approaches has the potential to improve evidence‐informed exercise prescription decision‐making.…”

“…The current study demonstrated how recently proposed statistical approaches (Hopkins 2000b;Swinton et al 2018) can be used to improve the confidence when classifying non-responders. The major novel findings are: 1) despite its growing popularity in the individual response literature (Scharhag-Rosenberger et al 2012;Astorino and Schubert 2014;Ross et al 2015;Gurd et al 2016;Alvarez et al 2017;de Lannoy et al 2017;Montero and Lundby 2017;Astorino TA et al 2018), threshold-based dichotomous classification approaches misclassify individuals who have a high probability of a positive response as non-responders, and 2) using response CIs with zeroand/or SWC-based thresholds provides more information than dichotomous classification approaches regarding an individual's response. Application of these statistical approaches has the potential to improve evidenceinformed exercise prescription decision-making.…”

“…; Astorino TA et al. ), which fails to consider “non‐response” within the context of meaningful benefit (Swinton et al. ).…”

“…Much of the recent literature has dichotomously classified individuals as either "responders" or "non-responders" using a pre-determined threshold (Scharhag-Rosenberger et al 2012;Astorino and Schubert 2014;Ross et al 2015;Montero and Lundby 2017) including two times the typical error (2x TE) (Bouchard et al 2012;Bonafiglia et al 2016;Gurd et al 2016;Raleigh et al 2016;Alvarez et al 2017;de Lannoy et al 2017;Astorino TA et al 2018). Although 2x TE is a relatively robust threshold for the classification of "responders" (i.e., the true change for observed responses >29 TE above zero is >90% likely to be positive) (Hopkins 2000a), dichotomous classification of "non-responders" has significant limitations.…”

“…Although 2x TE is a relatively robust threshold for the classification of "responders" (i.e., the true change for observed responses >29 TE above zero is >90% likely to be positive) (Hopkins 2000a), dichotomous classification of "non-responders" has significant limitations. First, dichotomous classification fails to consider the continuous range of probabilities of a positive response and almost certainly misclassifies individuals as "non-responders;" and second, 2x TE has typically been applied relative to zero Gurd et al 2016;Raleigh et al 2016;Alvarez et al 2017;de Lannoy et al 2017;Astorino TA et al 2018), which fails to consider "non-response" within the context of meaningful benefit (Swinton et al 2018).…”