ABSTRACT Low‐volume sprint interval training (LVSIT) increases peak oxygen uptake (VO 2peak ) when performed three times a week for 6 weeks. Methodological and statistical concerns, however, constrain the veracity of prior findings. We therefore reassessed the VO 2peak response to LVSIT using a randomized controlled trial design to mitigate bias and augment reporting quality. A generative model of VO 2peak was constructed as a function of group, baseline VO 2peak , age, sex, height, and change in body mass. Simulation experiments using previous data estimated that n = 15/group would achieve 80% power to detect a difference of 1 metabolic equivalent (MET) with a credible interval (CrI) of ≤ 1‐MET. Insufficiently active young adults (22 ± 3 years, body mass index: 24 ± 4 kg m −2 , baseline VO 2peak : 33 ± 7 mL kg −1 min −1 ) were randomized to perform 6 weeks of thrice weekly LVSIT ( n = 17) or no exercise (CTL; n = 20). The LVSIT protocol involved 3 × 20‐s “all out” sprints over a 10‐min session of low‐intensity cycling. Bayesian generative multivariate modeling revealed that LVSIT increased absolute +325 mL min −1 (101–605) and relative VO 2peak +5.6 mL kg −1 min −1 (2.2–8.1) versus CTL. All but one LVSIT participant (94%) were deemed a responder (i.e., mean estimate ± 95% CrI for relative VO 2peak > 0). In contrast, four CTL participants (20%) met this criterion. LVSIT also improved time to exhaustion by +133 s (101–160) versus CTL. We unequivocally demonstrate that 6 weeks of thrice weekly LVSIT increased VO 2peak in insufficiently active young adults compared to no exercise. By incorporating a robust design that included preregistration, concealed allocation assignment, statistical best practices, and applied Bayesian methods, and open data‐sharing, this study addresses prior methodological critiques of similar previous work.
Renwick et al. (Fri,) studied this question.