Author(s): ASPE, R., RITCHE, E., MIDDLETON, L., TAYLOR, B., SWINTON, P., Institution: ROBERT GORDON UNIVERSITY, Country: UNITED KINGDOM, Abstract-ID: 2299

INTRODUCTION:
The 30-second Wingate anaerobic test (WAnT30) is commonly used to assess an individual’s capacity to generate power from anaerobic energy systems in a laboratory [1]. Whilst the WAnT30 has been shown to be valid and reliable [2], criticisms and limitations are apparent [3]. The use of modified Wingate anaerobic tests (WAnT) comprising 6-, 10-, or 20-second protocols have been proposed as more effective alternatives to the WAnT30 where peak power (PP) is the measure of interest [1,2,4]. Moreover, anaerobic test accessibility has improved with the development of commercially available Wattbikes that integrate a cycling ergometer with user-friendly software. The primary objective of this study was to assess criterion validity of the Wattbike Pro to measure PP. Comparisons were made between a 6-second test on the Wattbike Pro (PPT6) and the WAnT6 and WAnT30 tests performed on a laboratory ergometer. Where systematic bias was encountered, a second objective of the study was to quantify uncertainty in standard correction equations.
METHODS:
Thirty-five participants (males: n=30, 21.3 ± 1.6 yrs, 182.3 ± 8.4 cm, 83.9 ± 12.2 kg; Females: n=5, 23.1 ± 2.2 yrs, 168.5 ± 4.8 cm, 69.1. ± 4.8kg) completed two testing sessions in a randomised order on separate days. One testing session included the WAnT30 and the other included the PPT6 and WAnT6 performed in a random order. A Bayesian framework was used to compare group means and to conduct Bland-Altman analyses to assess criterion validity. Bayes factor (BF) with qualitative interpretation of strength of evidence was used to interpret difference in group means. Stability of correction equations were assessed by quantifying uncertainty in regression parameters and fitted values when regressing WAnT6 and WAnT30 on PPT6.
RESULTS:
Comparison of WAnT30 and WAnT6 identified “extreme evidence” for greater PP produced during WAnT6 (Difference0.5 = 33.9 [95%CrI: 10.2 to 57.8 W]; BF>100). Bland-Altman analyses identified similar overestimations for PPT6 relative to both WAnT6 and WAnT30. No heteroscedasticity was observed, but proportional biases with overestimations of ~115 W for those at the 0.25-quantile and ~200 W for those at the 0.75-quantile were found. Substantive uncertainty was identified in regression parameters of correction equations corresponding to plausible changes of up to ~60 to 80 W in fitted values.
CONCLUSION:
The findings suggest that where PP is of primary interest the WAnT6 should be performed. Where there is no access to a laboratory ergometer, PPT6 can be an appropriate substitute. However, comparing PPT6 with WAnT values, or attempting to predict WAnT values should be used with caution.