From Growth Spurt to Injury Risk
Paper: “Associations between maturation assessment methods and growth-related injury occurrence in a German Bundesliga youth football academy”
Author: Laukhardt et al.
Journal: Scientific Reports (2026).
Why two fourteen-year-olds in the same academy can have a completely different injury profile
In youth development, age is a convenient administrative category, but biologically speaking it is a flawed measure. Two players who turn fourteen on the same day can be in entirely different stages of their physical development. New research from a German Bundesliga academy shows that the way you measure that development determines whether you protect the right player at the right time.
The study by Laukhardt and colleagues, published in Scientific Reports, compared nine different methods for determining biological maturation and peak height velocity (PHV) in 205 male youth players from U9 to U16. The players were measured four times over the course of a year, yielding 619 assessment-quarters for analysis.
The central question was remarkably practical: if different maturation methods identify different “risk periods”, which method best corresponds to the moment when growth-related injuries actually occur?
One PHV is not the same as another
In practice, a wide range of methods is used to estimate maturation: the anthropometric equations of Mirwald, Moore and Fransen (which predict the age at PHV), and approaches based on the percentage of predicted adult height, such as Khamis-Roche and the ultrasonic bone maturation method BAUSport.
Every method has to draw a line somewhere: pre-PHV, PHV or post-PHV. In the study, these boundaries are referred to as Red-Flag ranges. But that word matters: a Red-Flag range is not a biological switch. The authors explicitly emphasise that these windows are operational definitions derived from the literature, practice or – in the case of BAUSport – the manufacturer’s specifications. They are therefore not proven biological boundaries within which injury risk suddenly flips.
What the researchers found
In total, 85 non-contact injuries, 59 contact injuries and 31 growth-related injuries were recorded, alongside 444 injury-free assessment-quarters. The average time lost to a growth-related injury was 37.8 days – higher than for non-contact injuries (22.6 days) and contact injuries (21.4 days).
The ultrasonic BAUSport method showed the most consistent pattern. When players were in the PHV period according to BAUSport, the odds of a growth-related injury were higher than outside it: pre-PHV OR 0.33 and post-PHV OR 0.17 relative to PHV.
But here is a crucial nuance. BAUSport was not the only method to show an association. The long Khamis-Roche variant (OR 0.36) and the long Moore variant (OR 0.34) also showed significantly lower odds in the pre-PHV period relative to PHV. The difference: these two methods showed lower odds before PHV, but not significantly lower odds after PHV. BAUSport was the only method in which both the pre- and post-PHV periods scored significantly lower than the PHV period.
The study therefore does not say that BAUSport predicts injuries and all other methods are worthless. It says, rather: of the methods examined, BAUSport showed the most consistent pattern in this group.
The surprising story of the non-contact injury
Perhaps even more interesting for performance practice is what happened with the non-contact injuries. Across almost all maturation methods, the odds of a non-contact injury were found to be lower in the pre-PHV phase than in the PHV phase, after which the odds increased further in the post-PHV phase.
This emphatically does not mean that maturation itself is the cause of muscle or ligament injuries. The study was not designed for that. All kinds of factors can change simultaneously as a player gets older: training intensity, match level, sprint load, playing minutes and greater absolute forces. The researchers themselves emphasise that future research must account for previous injuries, actual growth rate and exposure to training and matches.
For contact injuries, no significant difference was found between the three maturation phases. That is logical: a knee injury after a duel is largely determined by an external event, not by internal biological vulnerability. PHV is therefore not a universal injury predictor – it is mainly relevant for a specific type of problem.
The biggest limitation: no exposure data
For football practice, this is perhaps the most important caveat. The study looks at injury incidence per assessment-quarter, but does not correct for how much football a player actually played. The authors acknowledge this explicitly: due to the lack of precise data on individual exposure time in training and matches, it was not possible to calculate injury incidence per 1,000 hours.
This is relevant, because older players generally train and play more and more intensively – a factor that can contribute to the odds, especially for non-contact injuries. The authors therefore warn that the associations found should be interpreted with caution.
In addition, chronological age and maturation status were strongly correlated in this cohort, which makes it difficult to distinguish the two from each other.
BAUSport: promising, but not yet a gold standard
The results make BAUSport interesting, but not yet the new gold standard. The research was conducted within a single Bundesliga academy, involved exclusively male players from U9 to U16, recorded only 31 growth-related injuries, and there was no independent gold standard for establishing the actual PHV.
There is more: the BAUSport Red-Flag range of 85–96% of predicted adult height is based on the manufacturer’s specification and has not yet been empirically validated as a biological boundary. That makes the results interesting, but also a reason to be cautious about commercial implementation.
The core message
The most important lesson is perhaps simpler than the statistics suggest:
Maturation is not an injury predictor with an on/off switch. It is context.
Whoever uses that context well can better understand why two players of the same calendar age can respond completely differently to the same training load. For the modern football academy, that means a shift from “How old is this player?” to: “Where does this player stand biologically, how much load is he receiving, how does he respond to it, and what signals is his body giving?”
That is precisely where the practical value of this research lies – not in replacing the coach, strength coach or physiotherapist with a maturation algorithm, but in giving better information to the people who ultimately have to make the individual development decision.
What should a youth academy do differently tomorrow?
- Measure structurally – make maturation not a one-off test, but a longitudinal process.
- Use one primary method consistently – know why you use a particular method and what uncertainties come with it.
- Record growth yourself – look not only at a predicted PHV date, but also at actual changes in height.
- Link maturation to exposure – a maturation profile without load information is incomplete.
- Make complaints visible before someone drops out – pain is information, not just an injury outcome.
- Use Red-Flag periods as a monitoring signal – not as an automatic ban on load.
- Make interventions individual – the combination of maturation, complaints, growth and load determines the next step.
- Evaluate your own population – a Bundesliga academy is not automatically representative of every Dutch youth academy.
- Avoid false certainty – an OR of 0.17 is not an individual prediction that a player has 83% less chance of injury.
- Keep development central – the goal is to let players train and play as healthily as possible through their development.
This article was produced with the assistance of artificial intelligence. All facts and statistics are drawn from the original peer-reviewed source, Laukhardt et al., Scientific Reports (2026)
