The study drew on data from a six-year, school-based physical activity initiative conducted in south-eastern Norway. Of 351 children whose parents provided consent, 330 completed at least one valid maximal treadmill test, an unusually rigorous approach in pediatric research where field estimates of fitness are far more common than direct measurement. Because valid peak oxygen uptake data from 2016 were available from only three of the nine participating schools following a technical data loss, the researchers based their inferential analyses on measurements collected in 2015, 2017, 2018, 2019 and 2020, providing a longitudinal window that spans most of primary school.

At the heart of the research lies a methodological question that has divided pediatric exercise scientists for decades: how should oxygen uptake be expressed when comparing children of different sizes? Raw absolute values in liters per minute naturally favor older, heavier children. Dividing by total body mass is the most common correction, but it penalizes children with higher body fat, who contribute mass but little metabolic machinery for running. Allometric scaling, which uses exponentials of body mass, and scaling to fat-free mass, which isolates metabolically active tissue, offer alternatives. The HOPP team, led by Asgeir Mamen of Kristiania University College, together with Julianna Buer of the Norwegian School of Sport Sciences and Per Morten Fredriksen of the University of Inland Norway, designated fat-free-mass-scaled peak oxygen uptake as their prespecified primary outcome and tracked all four expressions simultaneously.

The statistical architecture reflected the complexity of the design. Linear mixed models included test year, school allocation group, sex, and the interaction between test year and group as fixed effects, with participant identity as a random intercept and test year as the repeated factor. The primary model incorporated 904 observations from 314 children, a sample that gives the analysis substantial power to detect developmental trends. School-level models were run as sensitivity analyses to probe whether results were being driven by the clustering of children within individual schools, a known hazard of school-based interventions.

The headline result was unambiguous in direction, if nuanced in interpretation. Fat-free-mass-scaled peak oxygen uptake was significantly associated with test year, allocation group, sex, and the test year by group interaction, with all p-values at or below 0.030. From fifth to sixth grade, the adjusted mean fell by 5.73 milliliters per kilogram of fat-free mass per minute in the intervention group and by 5.11 in the control group. Critically, the difference between those group-specific changes was just 0.62 milliliters, with a 95 percent confidence interval spanning from minus 4.25 to 3.00 and a p-value of 0.736. In other words, the school-based physical activity program appeared to make no measurable difference to the final-year decline.

That equivalence between intervention and control schools is perhaps the study’s most sobering finding. The HOPP project was designed as a pedagogical experiment in which schools adopted enhanced physical activity programming, and the reasonable hope was that extra activity would buffer children against fitness losses as they approach adolescence. Instead, the data suggest the late-primary-school decline was resistant to the intervention as delivered. Body-mass-relative and allometrically scaled oxygen uptake also fell significantly in both groups during the final year, while absolute peak oxygen uptake, unadjusted for growth, held roughly steady, hinting that increases in raw aerobic capacity were simply failing to keep pace with children’s rapid physical development.

Sex differences emerged across every expression of fitness, with boys showing higher adjusted values than girls. Yet the magnitude of that gap depended heavily on how the data were scaled. When oxygen uptake was normalized to fat-free mass, which strips away the influence of differences in body composition between the sexes, the sex difference was substantially attenuated. This observation reinforces a growing consensus in pediatric physiology: apparent fitness gaps between boys and girls are partly artifacts of scaling choices and body composition rather than true differences in the oxidative capacity of muscle tissue.

The school-level sensitivity analyses added another layer of nuance. Changes in the three body-size-adjusted expressions of fitness were negative at all nine schools, suggesting the final-year decline was a pervasive phenomenon rather than a quirk of any particular classroom, neighborhood or teaching staff. Such consistency strengthens the biological plausibility of the trend, while also underscoring that whatever combination of growth, maturation and behavior drives the decline, it operates broadly across the study population and overwhelms school-level variation in programming.

The authors are candid about the limitations that temper firm causal conclusions. The study design could not separate growth-related developmental changes from shifts in behavior associated with the COVID-19 pandemic, which overlapped with the later measurement years and is known to have disrupted children’s physical activity worldwide, nor from the possibility that the composition of participants changed over time. A retrospective trial registration, dated 20 June 2015 on ClinicalTrials.gov as NCT02495714, and the 2016 data gap further complicate the inferential picture. Still, the convergence of negative findings across multiple scaling methods and all nine schools makes the late-primary decline difficult to dismiss as statistical noise.

For researchers, the message is methodological: conclusions about childhood aerobic fitness hinge materially on scaling method, body composition and school-level heterogeneity, and studies that report only a single expression of peak oxygen uptake may tell an incomplete or even misleading story. For parents, educators and policymakers, the findings add to mounting evidence that the transition out of primary school is a vulnerable window for children’s cardiovascular health, one that conventional school-based activity interventions, at least as implemented in HOPP, did not shield against. Identifying what does work during those years, and distinguishing genuine developmental biology from modifiable behavior, now stands as an urgent task for pediatric exercise science.

Subject of Research: Longitudinal development of peak oxygen uptake in 6- to 12-year-old children

Article Title: Longitudinal development of oxygen uptake in 6- to 12-year-old children: the Health Oriented Pedagogical Project (HOPP)

Article References: Mamen, A., Buer, J., & Fredriksen, P. M. (2026). Longitudinal development of oxygen uptake in 6- to 12-year-old children: the Health Oriented Pedagogical Project (HOPP). BMC Pediatrics. https://doi.org/10.1186/s12887-026-07631-7

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07631-7

Keywords: peak oxygen uptake, aerobic fitness, children, fat-free mass, allometric scaling, treadmill running, primary school, physical activity, longitudinal study, pediatric health, HOPP, Norway