Reply to Discussion of “Establishing modified Canadian Aerobic Fitness Test (mCAFT) cut-points to detect clustered cardiometabolic risk among Canadian children and youth aged 9 to 17 years” – The need for foundational fitness research in Canada: is there room for innovation?

Abstract

In their editorial, Tremblay and Tomkinson (2020) provided a critical reflection on the modified Canadian Aerobic Fitness Test (mCAFT) and its use in Canadian children and youth. They highlighted that (i) the validity and reliability evidence for the test is outdated in adults and nonexistent in children and youth aged <15 years; (ii) the background validation evidence for the test and prediction equation to estimate peak oxygen consumption (V O 2peak ) is often cited incorrectly in the literature; (iii) the Canadian Health Measures Survey (CHMS) exclusion criteria for the mCAFT were very strict and may bias the results; (iv) the mCAFT may not properly control for body weight when estimating V O 2peak ; and (v) using the mCAFT in research on temporal trends and international comparisons in cardiorespiratory fitness is difficult (Tremblay and Tomkinson 2020). While many of the issues raised have merit, we hope to clarify our position for publishing our mCAFT cut-points in Canadian children and youth (Lang et al. 2020). We also want to take this opportunity to highlight the expanding debate in the cardiorespiratory fitness literature which points toward a clear need for innovation and change.In our paper published in this issue (Lang et al. 2020) we established mCAFT cut-points to identify Canadian children and youth at increased risk of poor cardiometabolic health. We used established statistical methods in a large nationally representative sample of Canadians and obtained cut-point values that are considered high (males: 49-46 mL·kg −1 ·min −1 ; females 46-37 mL·kg −1 ·min −1 ) in comparison with previously identified international cut-points obtained using different field-based tests of cardiorespiratory fitness (males: ϳ42 mL·kg −1 ·min −1 ; females: ϳ35 mL·kg −1 ·min −1 ; Ruiz et al. 2016). These results were of concern to us. Furthermore, the mean mCAFT values obtained from the examined CHMS sample were substantially higher than values observed from the pooling of over one million 20-m shuttle-run test scores (Tomkinson et al. 2017). Upon further investigation our conclusions were similar to what Tremblay and Tomkinson concluded: the high mCAFT cut-points may have resulted from the exclusion criteria and/or the validity of the test.Although strict exclusion criteria holds merit, we were less concerned about this scenario given the fact that previous research involving the CHMS identified only small differences in body mass index between those included (23.2 kg·m −2 ) and excluded (24.1 kg·m −2 ) from the mCAFT (Tremblay et al. 2010). Slightly more problematic was the unresolved validity of the mCAFT in children aged <15 years. However, the importance of construct validity (i.e., the ability of the test to provide an accurate representation of what it purports to measure, in this case, V O 2peak in mL·kg −1 ·min −1 ) is relative to the objective of the research. For instance, it is very important to have strong construct validity (i.e., the ability to accurately predict V O 2peak ) when comparing results with ...