Comment on: Is high‐intensity interval training more effective on improving cardiometabolic risk and aerobic capacity than other forms of exercise in overweight and obese youth? A meta‐analysis

We commend the efforts of García-Hermoso and colleagues (1), whose recent meta-analysis concluded that highintensity interval training (HIIT) is more effective for improving systolic blood pressure (SBP) and maximal oxygen uptake (V̇O2max) than other forms of exercise in overweight and obese youth. However, while this is a timely article, we feel that the conclusions need to be contextualized with regards to important methodological constraints. Considering that this meta-analysis may shape health practitioner practice and public health policy, we strongly feel that these limitations should be addressed. The methodological constraints include the following: (i) limited number of trials – only five trials reported SBP, and only six reported V ̇O2max. Further, the authors admit that the quality of included trials was generally poor. (ii) Poorly characterized outcomes – for example, the six trials that reported V ̇O2max used a variety of measures, including differing exercise modalities and use of submaximal V ̇O2max estimation. (iii) Shifting baseline – the authors do not use a true control group, instead comparing the HIIT intervention to ‘other exercise’. The ‘other exercise’ groups vary widely in terms of exercise prescribed, with some ‘control’ groups arguably prescribed elements of HIIT, making it difficult to discern the true effect (and directionality) of HIIT prescription. Further, HIIT is loosely defined and dependent upon a seemingly infinite number of combinations of differing factors, including exercise modality, number of repetitions, work interval intensity and duration, among others. (iv) Inappropriate subgroup analysis – the authors do attempt to compensate for the lack of a true control group through sub-group analysis, comparing HIIT to different types of comparison exercise groups. However, only three trials were included in the V ̇O2max subgroup analysis. (v) Poor adherence to PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines (2) – for example, simple summary data should be reported for each intervention and control group to enable the reader to infer directionality. Further, additional analyses are poorly described; the meta-regression is poorly rationalized and difficult to interpret. The limitations of this study are alluded to but not explored. (vi) Poor participant inclusion criteria – inclusion of pre-adolescent and postadolescent children. A 6-year-old child will demonstrate a very different physiological response to exercise compared with a 17-year-old youth, largely due to age-dependent metabolic and musculoskeletal differences (3). Further, the authors reported that the weighted mean difference for SBP was 3.67mmHg, which is a greater relative change in a 6-year-old child with an expected resting SBP of ~96mmHg, compared with a 17-year-old adolescent with an expected resting SBP of ~118mmHg (4). Arguably more important than the methodological constraints is the lack of adequate consideration for clinical inference, implementation science and importance to public health. With regard to clinical inference, the weighted mean difference was 1.92mLkg min 1 for V ̇O2max and 3.67mmHg for SBP, both of which are within respective measurement error (5,6). As such, it is difficult to discern the clinical importance of such small changes. In terms of implementation and public health, HIIT may overcome the time barrier to exercise adherence, but there are a number of additional considerations that need to be addressed. Queries such as, but not limited to: should HIIT be tailored to the age and gender of the child? Will this form of exercise prescription put obese children at increased risk for musculoskeletal injury? Should some form of priming period precede the implementation of HIIT, i.e., should HIIT be employed following a period of strength training, which has been shown to improve cardiorespiratory fitness and psychological wellness in obese children, and is well-tolerated? (7,8) Will this form of exercise be enjoyable and sustainable? With regards to the last point, the authors cite one of our previous studies (9) as support for HIIT being more enjoyable compared to other forms of exercise. However, our previous study investigated a high intensity games intervention, not HIIT per se, and we did not compare the games intervention to other forms of exercise (9). Lastly, what kinds of physical resources and obesity reviews doi: 10.1111/obr.12451