Associations Between Bone Age, Body Composition and Physical Performance in Icelandic 12‐Year‐Old Female Soccer Players

ABSTRACT

Biological maturation significantly influences physical development and performance, with notable differences between sexes. Research using objective measures, such as bone age (BA) assessed with X‐ray to evaluate maturity and dual‐energy X‐ray absorptiometry (DXA) to assess body composition, remains scarce, particularly in females. This study investigated the associations between BA, body composition and physical performance in 12‐year‐old female soccer players. In total, 89 players (M_age 11.89 ± 0.33 years) from ten local soccer clubs across the greater Reykjavík area participated. BA, measured with X‐ray, indicated biological maturity, whereas DXA assessed body composition. Physical performance was measured with a 40‐m linear sprint, the countermovement jump (CMJ) and the Yo‐Yo intermittent recovery test (IR1‐test). Results showed that BA had a small negative correlation with 40‐m sprint time (r = −0.253, p = 0.017) but not with other performance tests. Total fat mass and fat percentage positively correlated with 40‐m sprint time (r = 0.351 and r = 0.566) and negatively with CMJ height (r = −0.534 and r = −0.632) and peak power (r = −0.490 and r = −0.636; all p < 0.001). Fat‐free mass and fat‐free mass index were negatively correlated with 40‐m sprint time (r = −0.299 and r = −0.301; both p = 0.004) but not with other tests. No body composition measures correlated with the IR1‐test. These findings emphasise the importance of understanding female‐specific maturation patterns and their impact on physical performance, highlighting the need for targeted research and tailored training programmes for youth female athletes.

Highlights

• Biological maturation plays a key role in physical performance, yet objective measures such as X‐ray and DXA are rarely used to study maturity and body composition in female adolescents, highlighting a critical research gap.

• Higher fat mass is associated with slower sprint times and lower jump performance, while greater fat‐free mass correlates with faster sprint times in female adolescents.

• The study highlights the importance of gender‐specific training programmes that prioritise fat‐free mass development and fat mass management during puberty, offering practical implications for youth training programmes aimed at optimising physical performance in female athletes.

1. Introduction

The physical changes associated with puberty are well‐documented (Malina 2004). However, there is limited research on how these changes specifically affect physical performance in females, as most studies have focused on mixed‐gender populations or exclusively on males. Addressing this gap is essential for developing training programmes that mitigate performance challenges during adolescence and reduce the risk of sports dropout. Assessing biological maturation is critical in understanding physical development and performance in youth athletes (Beunen and Malina 2008). During maturation, males and females differ in timing, rate and nature of growth due to influences from hormonal, genetic and physiological factors (Breehl and Caban 2018). In general, females mature earlier, reach their final height sooner, accumulate more fat mass and develop less muscle mass than males (Baxter‐Jones et al. 2008). In contrast, males experience a prolonged growth period, gain more muscle mass, and typically end up taller and more physically robust (Wells 2007). These differences in maturity trajectories have important implications for athletic performance (Beunen and Malina 2008), and may require different focus in the design of training programmes for male and female athletes (Lloyd, Oliver, et al. 2014). These differences could also influence sports participation, selection for academic teams, and opportunities in competitive activities and sports (Beunen and Malina 2008; Malina et al. 2000).

Chronological age (CA) is often used as a standard reference in research on growth‐related changes and performance (Lloyd, Oliver, et al. 2014). Nevertheless, there is considerable diversity in growth trajectories, maturation rates and performance abilities among individuals within the same CA, especially throughout puberty (Bergeron et al. 2015). Bone age (BA) has been identified as a reliable indicator of maturity status, suitable for assessing developmental progress from early childhood through late adolescence (Engebretsen et al. 2010; Malina 2011). Although skeletal maturation status is considered the gold standard for assessing biological maturity, few studies have examined the relationship of BA with physical performance in females. Most studies have relied on noninvasive measures and suggested that biological maturity minimally affects jumping, sprinting and intermittent endurance performance (Freitas et al. 2024; Malina et al. 2021).

Body composition, including factors such as fat mass (FM), fat‐free mass (FFM) and fat distribution, plays a critical role in determining physical performance by affecting strength, endurance, agility and overall athletic capability (Beunen and Malina 2008). FFM is positively associated with better performance in tasks requiring strength, speed and agility, while higher FM can negatively impact endurance and agility (Ben Mansour et al. 2021). Some studies have highlighted that consistent physical activity during adolescence has a positive effect on FFM gain, which is crucial for athletic performance (Maciejczyk et al. 2014).

When exploring previous data on the influence of body composition among females, methodological inconsistencies are evident, particularly in studies on youth athletes (Rodríguez et al. 2005). Most published data thus far have merely relied on skinfold measures, but dual‐energy X‐ray absorptiometry (DXA) is widely regarded as a reliable and accurate method for assessing body composition in athletic populations (Mathisen et al. 2023; Tornero‐Aguilera et al. 2022). Research assessing maturity levels and body composition of adolescents using objective measures, such as BA and DXA, is notably scarce. The gap in the literature regarding female development, particularly given the distinct physiological and developmental differences between the sexes during puberty, is particularly concerning when evaluating athletic performance. Female athletes experience unique growth and maturation patterns that can significantly influence physical capabilities and sports performance. Therefore, targeted research on female adolescents is essential to better understand their development and optimise performance outcomes, particularly in the context of athletic performance—an area currently underrepresented in the literature. To address these gaps in the literature, this study aimed to investigate the associations between BA, body composition and physical performance in 12‐year‐old female soccer players.

2. Materials and Methods

2.1. Study Design and Data Collection

This exploratory observational cohort study was conducted between April and June 2024 in Reykjavík, Iceland, targeting 12‐year‐old female soccer players. The study serves as the baseline data collection in a longitudinal project titled Soccer Knowledge for Optimal Resilience and Athletic Performance in Female Youth (the SKORA project), and uses cross‐sectional data to assess BA, body composition and physical capacities of players at the time of inclusion. Participants were from ten local clubs in the Reykjavík area, selected for their geographic diversity to represent the region and included players of varying levels. Soccer is the most popular sport in Iceland, with players typically training in mixed‐ability groups with their local clubs. Unlike many countries, Iceland does not separate grassroots and elite youth soccer, and clubs do not operate exclusive talent‐development academies. The women's national team has achieved significant success in international competitions. The country also boasts a robust coach education system, with all coaches required to complete training through the Icelandic Football Association (KSÍ), ensuring consistent coaching quality.

2.2. Participants

Participants were recruited through an introductory email sent by team coaches to players and their parents or guardians. Players were required to submit signed informed consent forms to enrol in the study. Eligibility was based on attendance at scheduled team practice sessions during designated measurement periods and parental consent. Measurements for most teams were completed during two practice sessions; however, for larger teams, data collection was extended to three or four sessions (only in the case of two teams). A total of 222 female soccer players, born in 2012, were registered within the ten teams, and 150 players chose to participate in the study, with a participation rate of 78.5%. Of the 150 players, 89 had valid data for X‐ray, DXA measures of body composition, sprint time, the countermovement jump (CMJ) and the Yo‐Yo intermittent recovery level 1 test (IR1‐test). Further details are shown in flowchart in Figure 1.

FIGURE 1.

Participants flowchart.

The National Bioethics Committee and the Icelandic Data Protection Authority approved the study (Study number: VSN‐24‐023), and the study was conducted in accordance with the guidance provided in the Declaration of Helsinki.

2.3. Skeletal Maturation

Players underwent an X‐ray of their left wrist to estimate BA based on skeletal maturity status. The X‐ray images were obtained at the Icelandic Heart Association using Siemens' Ysio Max with integrated imaging system FLUOROPSPOT Compact (software version VE10; Siemens Healthineers). The field of view covered the hand and 3 cm of the lower distal arm to include the epiphyseal plates in the radius and ulna. The parameters were as follows: tube‐detector distance 1 m, X‐ray energy 50 kV and 1–1.5 mAs, with no processing or filtering of the images. The radiographs were analysed using BoneXpert (standalone version 3.4.1.0; Visiana, Holte, Denmark), based on the Greulich and Pyle methodology (Thodberg et al. 2008). The system automatically calculates 8–13 independent BA measurements from different bones in the left hand, ruling out inter‐ and intra‐observer variability. The estimated root mean square error for BoneXpert is 0.68 years for males and 0.52 years for females (Maratova et al. 2023).

2.4. Body Composition

Body composition was measured using DXA with a LUNAR iDXA, Pro Advance scanner (GE Healthcare, Chicago, IL, US) at the Icelandic Heart Association by a certified radiologist. Participants arrived in a nonfasted state in the afternoon for the measurements. Standing height was measured with a stadiometer (seca 123, Seca Ltd., Birmingham, UK) to the nearest 0.1 cm. Body weight was measured to the nearest 0.1 kg using scales (seca 813, Seca Ltd., Birmingham, UK) with participants wearing light clothing. Body fat percentage was calculated by dividing FM by total body mass and multiplying the result by 100. FFM was calculated by subtracting FM from total body mass, and the fat‐free mass index (FFMI) was calculated by dividing FFM by the square of height in metres. The use of multiple body composition metrics enables a comprehensive analysis of the relationships between body composition and physical performance (Wells and Fewtrell 2006). FFM was selected as the primary measure over lean mass (LM) to ensure alignment with standardised definitions used in DXA‐derived assessments (Heymsfield et al. 2024). Although LM is sometimes used as a synonym for FFM, it may exclude bone mass in some contexts, depending on the measurement approach (Nana et al. 2015). Using FFM provides clarity and consistency in the analysis and facilitates comparison across studies employing similar methodologies.

2.5. Questionnaire

Participants completed a digital questionnaire at the team facilities for background information. To assess the players' maturation status, the question was asked: “Have you started menstruating (your period)?”. Response options were: 1 = “Yes”, 2 = “No”, 3 = “Don't know/prefer not to answer”. Participants were asked how long they had trained in soccer: “How old were you when you started playing soccer with a sports team?” Answer options ranged from age “4 and younger” and “up to 12 years”. Data on the frequency of soccer training sessions for each team were gathered from practice schedules, showing weekly sessions ranging from three to four per week, with each session lasting between 60 and 90 min.

2.6. Physical Performance Measures

Physical performance assessments and questionnaires were conducted at participants’ team facilities in April and May 2024. Participants adhered to the same standardised warm‐up protocol only before the IR‐1 and sprint test and the order of the tests was the same for all ten teams. The first testing day included an electronic questionnaire, CMJ, and the IR1‐test. The second day included the 40‐m linear sprint test. Because of time constraints during regular training sessions and the number of participants per team, it was not feasible to complete all tests within a single session. Therefore, the assessments were distributed over two days to ensure sufficient time for each test and to minimise participant fatigue. Players completed the running tests on artificial grass at the team's outdoor soccer field, but the CMJ test was performed at the team's indoor facilities.

2.6.1. Countermovement Jump Test

The CMJ test was performed to measure players' explosive lower body power in the vertical plane. As this was the first assessment of the day, no specific warm‐up protocol was performed prior to testing. Instead, players were familiarised with the jump technique and completed 2–3 practice jumps under the guidance of the research team, which served as preparation for the test. Players started in a standing position with their hands on their hips and legs extended. Players were instructed to jump as high as possible after an initial knee flexion. They were also instructed to straighten their legs in the air. The best of three attempts was used for the analysis. The CMJ test was performed with the ForceDecks (FDmini V.2, VALD, Brisbane, Australia). Jump height was calculated using the impulse‐momentum equation, which estimates height based on specific impulse metrics before take‐off (Xu et al. 2023). CMJ (W/kg) was calculated by dividing peak power by body weight.

2.6.2. Yo‐Yo Intermittent Recovery Test (IR1‐Test)

The Yo‐Yo Intermittent Recovery Level 1 test (IR‐1‐test) (Wood 2018) was used to assess players' intermittent endurance capacity, following original protocols (Bangsbo et al. 2008; KRUSTRUP et al. 2003). The test was preceded by a standardised 20‐min warm‐up, consisting of 10 min of low‐intensity running, 5 min of dynamic stretching, and three linear sprints of 40–50 m with progressive acceleration. During the IR1‐test, players performed repeated 2 × 20 m shuttle runs at increasing speeds, interspersed with 10 s of active recovery (slow jogging) between each shuttle. The test began at a velocity of 10.0 km·h⁻¹, with subsequent speed increments occurring according to a standardised audio signal. The audio signal, played via portable speakers, regulated the pace and ensured consistent timing across participants. The test continued until the player could no longer maintain the required pace. The total distance completed (in metres) was recorded for statistical analysis.

2.6.3. 40‐m Linear Sprint

A 40‐m linear sprint test was performed to measure the players' sprint speeds. All players followed a standardised 20‐min warm‐up protocol led by the measurement team before the test. The warm‐up consisted of a 10‐min low‐intensity run and 5 min of dynamic stretching. In the final part of the warm‐up, players performed 3 × 40–50‐m linear runs with increasing intensity and speed, followed by two maximal linear accelerations of 20 m. After the warm‐up, all players performed three maximal sprints of 40 m separated by 2–3 min of rest. The players started from a standing position with split legs, with the toes of the front foot placed 60 cm behind the first photogate, following the standardised sprint testing protocol used by Olympiatoppen, Norway's national high‐performance sports centre (Olympiatoppen 2024). The fastest sprint of three attempts was included in the analysis (Haugen and Buchheit 2016). A portable photogate system –Witty (Microgate, Bolzano, Italy)—measured the time between the gates. The height of the first photogate was 50 cm above the running surface, whereas the photogates at 20, 30 and 40 m were positioned 120 cm above the running surface (Haugen and Buchheit 2016).

2.7. Statistical Analysis

Statistical analyses were performed using R version 4.3.2 (v 4.3.2; R Core Team 2023) within the RStudio integrated development environment (version 2024.04.2+764; RStudio Team 2024). Descriptive statistics were calculated as means, standard deviations (SD), and ranges for all variables. The Shapiro–Wilk test was used to assess the normality of each variable and to guide the choice between Pearson or Spearman correlation methods. Pairwise associations between variables were examined using Pearson's product‐moment correlation coefficient (r) when both variables were normally distributed, and Spearman's rank‐order correlation coefficient (ρ) otherwise. For each correlation coefficient, p‐value, 95% and confidence interval (CI) and standard error (SE) were reported. For Spearman correlations, bootstrap resampling with 1000 iterations was used to estimate CI values in the correlation table. The strength of correlation coefficients was interpreted using conventional thresholds: value between 0.01 and 0.29 was defined as a small correlation, between 0.30 and 0.49 as a medium correlation, and from 0.5 to 1.0 as a large correlation (Cohen 1988). Statistical significance was set at p < 0.05. Correlation plots were created to visually illustrate the associations between variables. These plots display scatter distributions, trend lines, and annotated correlation coefficients with corresponding p‐values. Unlike the table, the plots were intended for visualisation only and did not include bootstrapped confidence intervals.

3. Results

Data from 89 participants were included. Most participants (72.6%) started organised soccer before the age of seven and on average, organised soccer training sessions were 3.75 ± 0.48 h per week. Only two players (2.2%) reported menarche, and eight (9%) did not answer the question. Physical performance characteristics and test scores are summarised in Table 1. The players' average CA and BA were 11.89 ± 0.33 and 11.77 ± 1.09 years, respectively. The range of BA, representing the maturity of the girls' skeletal systems, had a wider range than CA.

TABLE 1.

Overview of the participants' characteristics and scores from the physical performance tests.

Characteristics (n = 89)	Mean ± SD	Range (min–max)
CA (years)	11.89 ± 0.33	11.43–12.43
BA (years)	11.77 ± 1.09	8.80–14.57
Height (cm)	154.3 ± 7.2	136.0–169.8
Weight (kg)	44.44 ± 8.36	28.80–72.00
FM (kg)	12.15 ± 4.56	6.47–31.15
BF (%)	26.79 ± 5.83	14.47–43.26
FFM (kg)	32.25 ± 5.05	21.09–46.81
FFMI (kg/m2)	13.5 ± 1.26	10.9–17.0
Performance tests (n = 89)
20 m sprint (sec)	3.73 ± 0.23	3.28–4.43
30 m sprint (sec)	5.33 ± 0.33	4.70–6.41
40 m sprint (sec)	6.98 ± 0.46	6.10–8.48
CMJ (cm)	19.39 ± 3.68	11.9–29.5
CMJ (W/kg)	35.96 ± 4.90	25.90–48.00
IR1‐test (m)	342.92 ± 168.29	80–840

Abbreviations: BA, bone age: BF, body fat; CA, chronological age; CMJ, countermovement jump; FFM, fat‐free mass; FFMI, fat‐free mass index; FM, fat mass; IR1‐test, Yo‐Yo intermittent recovery test.

3.1. Correlations Between Bone Age, Body Composition and Physical Performance

There was a small significant negative correlation between BA and the 40‐m sprint time (r = −0.253, p = 0.017) but no correlations were found between BA and the other performance tests (Table 2). Total FM and fat percentage showed a positive correlation with 40‐m sprint time (ρ = 0.334 and ρ = 0.566), but a negative correlation was found with CMJ height (ρ = −0.513 and ρ = −0.632) and CMJ peak power (ρ = −0.485 and ρ = −0.624; all p < 0.001). FFM and FFMI were negatively correlated with 40 m sprint (r = −0.299 and r = −0.301; p = 0.005 and 0.004), but not any of the other tests. No correlations were found between any of the body composition measures and the IR1‐test. Correlation plots of the associations between biological maturity levels, body composition and physical performance are shown in Supporting Information S1: Figure a–c.

TABLE 2.

Associations between biological maturity levels, body composition and physical performance.

	40 m sprint	CMJ (cm)	CMJ (W)	IR1‐test (m)	BA (years)
BA (years)	r = −0.253	r = −0.009	r = 0.086	ρ = −0.018
	p = 0.017	p = 0.932	p = 0.423	p = 0.346
	[−0.438, −0.047]	[−0.217, 0.200]	[−0.125, 0.289]	[−0.309, 0.111]
	SE = 0.105	SE = 0.107	SE = 0.106	SE = 0.105
FM (kg)	ρ = 0.334	ρ = −0.513	ρ = −0.485	ρ = −0.105	ρ = 0.398
	p < 0.001	p < 0.001	p < 0.001	p = 0.327	p < 0.001
	[0.136, 0.513]	[−0.663, −0.329]	[−0.658, −0.276]	[−0.318, 0.115]	[0.198, 0.562]
	SE = 0.098	SE = 0.086	SE = 0.101	SE = 0.107	SE = 0.009
Fat%	ρ = 0.566	ρ = −0.632	ρ = −0.624	ρ = −0.055	ρ = 0.018
	p < 0.001	p < 0.001	p < 0.001	p = 0.608	p = 0.865
	[0.398, 0.698]	[−0.750, −0.474]	[−0.769, −0.463]	[−0.293, 0.163]	[−0.168, 0.226]
	SE = 0.076	SE = 0.007	SE = 0.083	SE = 0.117	SE = 0.100
FFM (kg)	r = −0.299	r = 0.03	r = 0.12	ρ = −0.130	r = 0.782
	p = 0.005	p = 0.783	p = 0.263	p = 0.223	p < 0.001
	[−0.477, −0.096]	[−0.180, 0.236]	[−0.091, 0.230]	[−0.329, 0.889]	[0.685, 0.851]
	SE = 0.098	SE = 0.107	SE = 0.106	SE = 0.106	SE = 0.042
FFMI (kg/m2)	r = −0.301	r = 0.105	r = 0.166	ρ = −0.0006	r = 0.641
	p = 0.004	p = 0.328	p = 0.121	p = 0.952	p < 0.001
	[−0.480, −0.099]	[−0.106, 0.307]	[−0.044, 0.361]	[−0.226, 0.213]	[0.499, 0.749]
	SE = 0.098	SE = 0.106	SE = 0.104	SE = 0.108	SE = 0.063

Note: Pearson's correlation coefficient (r) was used for normally distributed data, otherwise Spearman's rho (ρ) was used. Confidence intervals (95%) for each coefficient are presented in brackets and standard error as SE. Bold values indicate statistical significance.

Abbreviations: BA, bone age; BF, body fat; CA, chronological age; CMJ, countermovement jump; FFM, fat free mass; FFMI, fat free mass index; FM, fat mass.

4. Discussion

The purpose of this study was to explore the associations between BA, body composition and physical performance in 12‐year‐old female soccer players. Results showed that girls with higher BA had faster 40‐m sprint times. However, no significant correlations were found between BA and the other physical performance tests. Body composition also played a role in performance outcomes, total FM and fat percentage were positively correlated with 40‐m sprint time and negatively correlated with CMJ height and peak power. Furthermore, FFM and FFMI were associated with faster 40‐m sprint times, although FFM did not correlate with any other performance tests. No significant findings were observed between any body composition measures and the IR1‐test.

The data revealed a negative correlation between BA and sprint times, indicating that as BA increased, sprint times decreased—meaning that athletes who were more biologically mature tended to run faster. However, no significant associations were found with other performance tests. These findings are somewhat consistent with Gundersen et al. (2024), who also investigated BA using X‐ray in female athletes with a mean age of 14.0 years. In their study, participants performed a 40‐m sprint, standing long jump, push‐ups, and a 2000‐m run, and no significant relationships between BA and physical performance were found. However, Gundersen et al. (2024) explored the relationship in older girls with a smaller sample size and found that the 40‐m sprint test was close to significant (p = 0.073) (Gundersen et al. 2024). This suggests that age, sample size and other methodological differences may influence the observed relationships between BA and physical performance. Studies investigating the relationship between BA and performance in youth female populations are limited. Studies employing anthropometric equations to estimate peak height velocity (PHV) have reported significant relationships between maturity and performance in tasks such as vertical jumping (Almeida‐Neto, de Medeiros, et al. 2021a; Ried‐Larsen et al. 2015) and standing long jumps (Detanico and Kons 2023) among young female athletes aged 7–17 years old. Furthermore, a recent study by Runacres et al. (2023) reported a significant relationship between maturity, measured using PHV and 30‐m sprint performance among females aged 11–16 (Runacres et al. 2023). These findings indicate that more biologically mature females tend to perform better in jumping and sprinting tasks, which partly contradicts the results of this study, where more biologically mature players ran faster in the 40‐m sprint test, although no associations were found with the jump or endurance tests.

This discrepancy may stem from differences in maturity assessment methods, as PHV estimations in females are less precise, whereas BA provides a more accurate measure of biological maturity (Engebretsen et al. 2010). Although neuromuscular adaptations during growth and maturation likely enhance sprinting and jumping performance (Radnor et al. 2018), increased FM during adolescence may counteract these improvements in weight‐bearing activities. Longitudinal studies suggest that jump and sprint performance in females typically peaks around ages 13–14 before stabilising (Tingelstad et al. 2023; Tønnessen et al. 2015). Similarly, maximal oxygen uptake relative to body mass has been shown to decline as females progress through adolescence (Almeida‐Neto, Oliveira, et al. 2021b; Ingvarsdottir et al. 2024).

In this study, body composition emerged as a more significant determinant of athletic performance than BA. FM and fat percentage were correlated with slower 40‐m sprint times and lower CMJ height and peak power, indicating that higher fat levels may impair sprinting performance. This suggests that greater body fat may negatively affect sprint performance, possibly due to increased body weight and reduced power‐to‐weight ratio. However, higher muscle mass was associated with faster sprinting, as players with higher FFM tended to run faster (Almeida‐Neto, de Medeiros, et al. 2021a). This suggests that greater muscle mass may contribute to improved sprint performance, likely due to increased strength and power output. The results align with those of Granacher et al. (2016), who emphasised the importance of FFM in enhancing strength and power in young athletes (Granacher et al. 2016). Similarly, Malina et al. (2004) found that body composition, particularly FFM, is a critical factor in athletic performance, especially in female athletes (Malina et al. 2004). Higher FFM, particularly skeletal muscle, is a primary determinant of strength, speed and agility, reinforcing the need for strength and conditioning programmes that prioritise maximising FFM while effectively managing FM to optimise performance in female athletes (Pezoa‐Fuentes et al. 2023; Nuzzo and Pinto 2024). Therefore, training programmes should focus on maximising FFM (particularly muscle mass) while managing FM to improve female performance in strength, speed and agility tasks. However, a 2016 scoping review highlighted the lack of research on the sex‐specific effects of resistance training in females—a limitation that needs to be addressed in future studies (Granacher et al. 2016).

The lack of association between bone age, body composition, and IR1‐test performance may reflect the multifactorial demands of the IR1‐test itself, which incorporates aerobic and anaerobic components, rather than limitations in the players' aerobic capacity per se (Garcia‐Tabar et al. 2024). This nuance is particularly important in female youth populations, where variability in neuromuscular coordination, motivation, and pacing strategies may further influence intermittent test outcomes. Some previous studies with boys have demonstrated that those who mature earlier typically have greater muscle mass and often outperform their peers in physical performance metrics (Malina 2004; Philippaerts et al. 2006), while other studies with boys did not find a correlation (Grendstad et al. 2020). The lack of correlation in this study suggests that other factors, such as gender, effort, motivation or testing conditions, may play a more significant role in the IR1‐test performance in this female population. Trained soccer athletes have been reported to achieve greater distances in the IR1‐test than their age‐matched untrained peers. For instance, Póvoas et al. (2016) found that soccer‐trained children aged 9–11 covered 40% more distance than their untrained counterparts, with this difference increasing to 85% in players aged 12–13 years (Póvoas et al. 2016). Similarly, a 2018 British study examined elite youth female soccer players aged 10–16 and reported that older players achieved greater distances on the IR1‐test (Emmonds et al. 2018). However, neither study investigated the influence of BA or body composition on test performance, limiting the ability to make direct comparisons or explore underlying factors affecting performance. It is also likely that effort and motivation affected the IR1‐test performance. The test requires participants to push themselves to exhaustion and maintain consistent recovery intervals, making intrinsic motivation and psychological resilience critical determinants of success. These factors may overshadow physiological predictors, particularly in field settings where motivational levels and group dynamics vary significantly.

BA was significantly correlated with both FFM and FM, but not with fat percentage, with the strongest correlation observed between BA and FFM. This suggests that as skeletal maturity progresses, FFM increases correspondingly, highlighting a close relationship between bone development and muscle mass at this age (Tanner 1981). Additionally, although FM was also correlated with BA, the lack of association with fat percentage implies that the proportion of fat in the body remains relatively independent of skeletal maturation. This could be due to FFM and FM increasing simultaneously, keeping the overall fat percentage stable. These results suggest that BA is more strongly associated with absolute body composition measures rather than relative fat distribution, which has important implications for growth assessments and physical development evaluations, particularly in sports science and paediatric health.

Although elite benchmarks, such as those reported by Ruiz‐Rios et al. (2024), provide valuable context, direct comparison with our cohort of 12‐year‐old nonelite players is limited due to differences in biological maturity and training background. Similarly, although studies such as Emmonds et al. (2018) report youth performance data, they often reflect academy‐level populations. Our sample represents a grassroots setting in Iceland, which may offer a more typical reflection of youth development in nonelite environments (Emmonds et al. 2018; Ruiz‐Rios et al. 2024).

The study highlights the importance of gender‐specific training programmes that prioritise the development of FFM and the management of FM during puberty. This recommendation is supported by Lloyd, Faigenbaum, et al. (2014), who advocate for tailored training approaches to optimise performance and reduce injury risks in female athletes. Puberty brings significant physical and psychological changes, including shifts in body composition and physical capacity, which vary in timing and duration between individuals. These changes can affect self‐perception, motivation and performance, increasing the risk of dropout if not properly addressed. Coaches play a critical role in supporting athletes through this phase by normalising temporary performance declines and emphasising that these changes are part of normal development. Individualised support and fostering a team culture that validates and supports diverse pubertal experiences are essential. Educating coaches on the implications of puberty and designing training programmes tailored to individual development can enhance athlete retention, confidence and long‐term success. By addressing these challenges, organisations can help young female athletes thrive despite temporary disruptions in their athletic journey.

To our knowledge, this is the first study to examine the relationship between BA, body composition and physical performance in 12‐year‐old female soccer players using both X‐ray imaging and DXA scans to assess maturity and body composition precisely. Although the sample is homogenous, it represents Icelandic female youth soccer players from ten different clubs and neighbourhoods, strengthening the results. Additionally, the study employed validated and reliable field tests to measure physical performance, providing valuable insights into the factors influencing athletic development in this underrepresented demographic.

However, several limitations should be acknowledged. The sprint and endurance tests were conducted outdoors, which may have introduced variability in performance measures due to environmental factors. Efforts were made to minimise this by assessing weather conditions on test days and ensuring consistent and suitable testing environments across sessions. Although the IR1 test provides a soccer‐specific assessment of endurance (KRUSTRUP et al. 2003), it does not represent a pure measure of aerobic capacity (e.g., lactate threshold or VO₂ kinetics) and may be influenced by motivation, pacing strategy, and neuromuscular fatigue (Bangsbo et al. 2008; Garcia‐Tabar et al. 2024). Future studies may consider more controlled, submaximal tests of aerobic function where feasible. The study also lacked a direct measure of strength, which is an important component of physical performance. Including strength assessments in future testing phases would provide a more complete picture of athletic capacity in youth female athletes. Additionally, while the sample included players from ten different teams, exploratory analyses did not reveal systematic variation in performance by team. Visual inspection of key variables (e.g., CMJ, body fat %, IR1) showed no clustering, suggesting that observed variability was driven by individual factors such as developmental stage, multisport participation, or natural variation, rather than by team‐based routines. Lastly, the cross‐sectional design limits the ability to establish causality between maturity, body composition and performance outcomes. Future longitudinal research is essential to understand how these relationships develop through adolescence and to assess the impact of targeted interventions on the athletic development of female youth athletes. Such studies will help to build a more comprehensive understanding of female athletic progression during critical developmental periods.

5. Conclusion

This study underscores the critical role of biological maturation in physical performance, highlighting the need for objective measures, such as X‐ray and DXA, to assess maturity and body composition in female adolescents. Results suggest that higher FM is associated with slower sprint times and lower jump performance, whereas greater FFM and higher BA correlates with faster sprint times. These insights emphasise the importance of gender‐specific training programmes that focus on developing FFM and managing FM during puberty, providing practical implications for youth training programmes aimed at optimising physical performance in female athletes. Future studies should incorporate strength measures and longitudinal designs to track these relationships over time, while also exploring sex‐specific training strategies and the psychological factors influencing female athlete development.

Ethics Statement

The National Bioethics Committee and the Icelandic Data Protection Authority approved the study (Study number: VSN‐24‐023).

Consent

The participants provided their written informed consent to participate in this study.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1

Stefansdottir, Runa , Gundersen Hilde, Benediktsson Sigurdur, Vestbøstad Mona, Johannsson Erlingur, and Rognvaldsdottir Vaka. 2025. “Associations Between Bone Age, Body Composition and Physical Performance in Icelandic 12‐Year‐Old Female Soccer Players.” European Journal of Sport Science: e70029. 10.1002/ejsc.70029.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.