Nonlinear association between physical fitness and sports injury risk among Chinese high school students: a cross-sectional study

Abstract

Background

Sports injuries are a prevalent public health issue among adolescents, yet the association between physical fitness and injury risk in Chinese high school students remains under-investigated. This study aimed to evaluate the relationship between physical fitness levels and the risk of sports-related injuries among high school students in China.

Methods

A cross-sectional study was conducted among 1,659 students aged 15–18 years from six high schools in Shanghai. Physical fitness was assessed using standardized national tests (e.g., sprint, endurance run, standing long jump, sit-and-reach, strength measures). Sports injury data were collected via validated questionnaires. Students were categorized into quartiles based on composite fitness scores. Multivariate logistic regression models were used to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for injury risk across fitness levels. A two-piecewise linear regression was applied to explore potential non-linear associations.

Results

The overall injury prevalence was 23.0%, higher in boys (26.7%) than girls (19.2%). Compared to the lowest fitness quartile, students in the highest quartile had a significantly lower injury risk (adjusted OR = 0.13, 95% CI: 0.08–0.17). A clear dose–response relationship was observed (P for trend < 0.001). Below a composite fitness score of 65, each 1-point increase was associated with a 14% reduction in injury odds (OR = 0.86, 95% CI: 0.83–0.89); above this threshold, the association was no longer significant.

Conclusions

Physical fitness was inversely associated with sports injury risk among Chinese high school students. A threshold effect was identified at a fitness score of 65, below which injury risk declined sharply with each point increase. Interventions should prioritize students below this cutoff to effectively reduce injury incidence.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-25280-w.

Introduction

Sports injury is commonly defined as any injury occurring during sports or exercise that leads to an interruption of training or competition [1]. The high school period is a critical stage of development from adolescence to adulthood, and health status during this phase can profoundly affect an individual’s future physical growth, mental well-being, and social adaptation [2]. Regular physical activity plays an irreplaceable role in promoting adolescent health and preventing various diseases; for example, exercise in youth is associated with lower risks of heart disease, diabetes, and osteoporosis, as well as better mental health and cognitive function [3–5]. However, the potential risk of sports injuries cannot be ignored, as it has become a key factor discouraging students from active sports participation. Epidemiological studies report that roughly 10–25% of high school students experience sports-related injuries [6]. Such injuries not only curtail students’ engagement in physical activities but also adversely impact their daily academic life and create significant economic and psychological burdens on families. In the United States, for instance, high school sports injuries account for an estimated 2 million injuries and 30,000 hospitalizations each year, underscoring the substantial individual and societal burden [7].

Physical fitness refers to the comprehensive capacity of the body’s functions and structures, developed under genetic and environmental influences through long-term training. It encompasses various components, including cardiorespiratory endurance, muscular strength, flexibility, and other attributes [8]. The overall physical fitness of adolescents has shown concerning trends in recent years. Large-scale national surveys in China have documented a declining trend in youth fitness levels over the past few decades, with relatively few teenagers meeting desirable fitness standards. For example, one regional study found that only about 10–15% of high school students achieved “good” or “excellent” fitness ratings in annual evaluations [9]. Physical fitness is thought to have a substantial impact on sports injury risk. Adolescents with higher fitness levels generally demonstrate better neuromuscular control, strength, and endurance, which can enhance joint stability and resilience to stress, potentially reducing injury risk. Conversely, poor fitness or deficiencies in certain fitness components may leave individuals more susceptible to sprains, strains, and other injuries [10]. Indeed, empirical evidence indicates that low performance in fitness tests (such as slow endurance run times or weak muscular strength) is associated with a higher incidence of musculoskeletal injuries [11]. This suggests that improving physical fitness might be a viable strategy to mitigate sports injuries in the youth population.

Existing research supports the link between physical fitness and injury risk, though most studies have focused on college students or athletic subgroups. In university populations, higher physical fitness has been significantly associated with a lower risk of sports injuries, whereas those with poorer fitness tend to incur injuries more frequently [12]. Domestic studies in China likewise have identified multiple fitness-related risk factors for sports injuries. Characteristics such as age and sex, as well as body composition and specific fitness indicators, appear to influence injury occurrence. For instance, adolescents with higher body mass index (overweight/obesity) have been found to have elevated injury risk compared to their normal-weight peers. Lower levels of fundamental physical qualities—including reduced cardiopulmonary fitness (e.g. low vital capacity), weaker muscular strength (e.g. grip strength), and poorer explosive power (e.g. vertical jump performance)—have also been implicated as significant risk factors for injury in youth athletes [13, 14]. However, despite these insights, research on the relationship between physical fitness and sports injuries specifically in high school students remains relatively scarce. Many prior studies, especially in China, have concentrated on college students or other age groups, and systematic investigations targeting the high school demographic are still lacking. This gap highlights the need for focused research on high school populations to better understand how fitness level impacts injury risk in this critical age group.

This study aims to systematically evaluate physical fitness levels among high school students and examine their association with the risk of sports injury. By analysing how different components of physical fitness relate to injury occurrence, the study seeks to clarify the underlying mechanisms linking fitness and injury risk. The findings will provide evidence-based guidance for developing targeted injury prevention strategies and inform school sports administrators and educational policymakers on effective ways to improve students' physical fitness and ensure safer sports participation.

Methods

Study design

This cross-sectional study was conducted from September to November 2022 to investigate the association between physical fitness and the risk of sports injuries among high school students in Shanghai, China. Stratified cluster random sampling was employed to recruit participants from six high schools and secondary vocational schools. Students aged 15–18 years from grades one to three were selected, with each grade consisting of 120 students (50% boys, 50% girls), totalling 2160 students. Participants completed a validated comprehensive questionnaire designed to evaluate sports injury prevention and collect demographic data. Additionally, on-site physical fitness assessments were conducted.

The study protocol was reviewed and approved by the Ethics Committee of Shanghai Qunxing Vocational and Technical School (Approval number: 202004065). Prior to data collection, detailed information regarding the study's purpose, procedures, and potential risks was communicated to school principals, teachers, students, and their guardians. Informed consent was obtained from all participants and their legal guardians, and participation was entirely voluntary.

Participants

A total of 2160 students were initially recruited. Inclusion criteria required students to be aged 15–18 years, currently enrolled in the selected schools, and physically able to participate in regular school-based physical education and sports activities. Exclusion criteria included severe physical disabilities, chronic illnesses that precluded physical activity, or cognitive impairments. After applying these criteria, 1978 students met the inclusion standards. A total of 1978 questionnaires were distributed, with 1869 completed questionnaires returned (response rate: 94.5%). After excluding 46 questionnaires due to irregular and invalid responses, 1823 valid questionnaires remained (valid response rate: 97.5%). Physical fitness assessments were completed by 1798 students (participation rate: 90.9%). After removing incomplete or invalid physical fitness data (n = 79), the final valid data from physical fitness tests was 1719 (effective rate: 95.6%). All included students participated in regular school physical education, and most also engaged in extracurricular sports activities, although type and frequency varied. Finally, matching questionnaire and physical fitness data yielded a combined analytical sample of 1659 students (846 boys, 50.9%; 813 girls, 49.1%).

Procedures and measurements

Before data collection, comprehensive training sessions were conducted for all researchers, who were postgraduate students specializing in sports sciences. Researchers explained the study objectives and procedures to participants in classroom settings. Under supervision, each student completed the “School Sports Injury Prevention” questionnaire, previously published by Hu et al. [15], within 20 min in a classroom environment. The questionnaire collected detailed information about sports injury occurrences and habitual physical activity, including duration, frequency, and intensity. Consistent with the questionnaire instructions, injury recall was restricted to events occurring within the past year that required medical treatment. Specific details of the questionnaire’s validity and reliability have been previously reported [15]. In addition, the complete questionnaire content has been made available as Supplementary File 1 to facilitate replication and critical appraisal.

Following the completion of the questionnaire, students underwent a standardized physical fitness assessment, which was administered by trained research assistants in accordance with the 2014 revised National Student Physical Health Standards [16]. The assessment comprised body morphology (height, weight) and multiple fitness indicators, including the 50 m sprint, standing long jump, sit-and-reach, pull-ups [boys], sit-ups [girls], and the 1000 m run [boys]/800 m run [girls]. Scores were calculated based on gender-specific performance in each physical fitness indicator, and overall physical fitness scores were determined using percentile ranks and categorized into four quartile groups.

To ensure data integrity and accuracy, a rigorous double-entry method was implemented, where two trained research assistants independently entered data into a secure database with thorough cross-validation checks. Questionnaire responses and physical fitness data were anonymized using unique identifiers, and database access was strictly limited to authorized personnel to safeguard participants' privacy.

Statistical analysis

Participants were stratified into quartiles (Q1: 0–55.2, Q2: 55.3–65.2, Q3: 65.3–75.4, Q4: 75.5–100) based on their overall physical fitness scores. Logistic regression analyses were employed, using the lowest fitness quartile (Q1) as the reference group, to evaluate the odds ratios (OR) and 95% confidence intervals (CI) for sports injury occurrences across different fitness quartiles. Tests for linear trends across quartiles (P for trend) were performed. Multivariate linear regression analyses were conducted to explore the linear relationship between physical fitness scores and sports injury incidents. Additionally, segmented linear regression models were used to examine potential nonlinear relationships between these variables.

Model 1 was unadjusted. Model 2 was adjusted for sex and grade, as these demographic factors are well-documented determinants of both physical fitness performance and sports injury susceptibility. Model 3 was further adjusted for weekly exercise duration, exercise frequency, and exercise period. These exercise-related variables were included to account for total physical activity exposure, which represents an essential confounder. Students with higher levels of activity are more likely both to achieve superior physical fitness and to be at increased risk of sustaining injuries due to greater exposure opportunities. Adjusting for these factors allows the models to better isolate the independent effect of physical fitness on injury risk. Statistical analyses were performed using EmpowerStats software within the R statistical environment. Statistical significance was set at α = 0.05..

Results

Participant characteristics and basic information

A total of 1,659 high school students (846 males and 813 females) were included in the analysis after merging the valid questionnaire and physical fitness test data. The mean age of the participants was 16.23 ± 1.20 years, with a mean height of 169.9 ± 6.5 cm, mean weight of 65.8 ± 11.6 kg, and mean body mass index (BMI) of 22.58 ± 5.31 kg/m². The overall incidence of sports injury among the students was 23.0% (382/1659). Male students had a higher injury rate (26.7%, 226/846) compared to female students (19.2%, 156/813). Table 1 summarizes the distribution of sex, school type, and sports injury occurrence across quartiles of physical fitness test scores (Q1–Q4). Notably, female students constituted a larger proportion of the highest fitness group (64.2% in Q4) than the lowest fitness group (35.5% in Q1), and students from vocational high schools were more represented in the upper fitness quartiles (52.8% in Q4 vs. 36.7% in Q1). In addition, the proportion of students who experienced sports injuries was smaller among those with higher physical fitness levels than among those with lower physical fitness levels: 31.3% of students in the lowest fitness quartile (Q1) had a history of sports injury, compared to only 14.8% in the highest quartile (Q4).

Table 1.

Distribution of gender, school type, and sports injury incidence across physical fitness quartiles among high school students (n = 1659)

Physical Fitness Quartile	n	Gender, n (%)		School Type, n (%)		Sports Injury, n (%)
		Male	Female	Vocational	High school	Yes	No
Q1 (lowest)	419	270 (64.5)	149 (35.5)	154 (36.7)	265 (63.3)	131 (31.3)	288 (68.7)
Q2	412	239 (38.0)	173 (42.0)	175 (42.4)	237 (57.6)	109 (26.4)	303 (73.6)
Q3	415	189 (45.6)	226 (54.4)	197 (47.5)	218 (52.5)	81 (19.6)	334 (80.4)
Q4 (highest)	413	148 (35.8)	265 (64.2)	218 (52.8)	195 (47.2)	61 (14.8)	352 (85.2)

Association between physical fitness and sports injury risk

Logistic regression analysis was performed to examine the association between physical fitness levels and the risk of sports injury. Table 2 presents the odds ratios (OR) and 95% confidence intervals (CI) for sports injury across the physical fitness score quartiles, using the lowest quartile (Q1) as the reference group. In the unadjusted model (Model 1), higher physical fitness was associated with substantially lower odds of sports injury. Specifically, the unadjusted ORs for sports injury in Q2, Q3, and Q4 were 0.43, 0.32, and 0.23, respectively (all P < 0.05 vs. Q1), showing that students with better physical fitness had significantly reduced injury risk compared to those in the lowest fitness group. After adjusting for sex and grade (Model 2), the association became even stronger; for instance, students in the highest fitness quartile (Q4) had an OR of 0.16 (95% CI: 0.11–0.22) relative to Q1. In the fully adjusted model (Model 3), which controlled for sex, grade, weekly exercise duration, exercise frequency, and exercise period, higher fitness levels remained a significant protective factor against sports injuries. Students in Q4 had about 87% lower odds of experiencing a sports injury compared to those in Q1 (OR = 0.13, 95% CI: 0.08–0.17). Similarly, students in Q2 (OR = 0.39, 95% CI: 0.28–0.61) and Q3 (OR = 0.25, 95% CI: 0.19–0.34) also showed significantly reduced odds of injury relative to Q1 in the fully adjusted analysis (all P < 0.05). There was a clear dose–response relationship across the fitness quartiles, with each increment in fitness group associated with a further reduction in injury risk (P for trend < 0.01).

Table 2.

Association between physical fitness quartiles and sports injury risk among high school students: results from logistic regression models (OR [95% CI])

Fitness Quartile	Total (n)	Injured (n)	Model 1 OR (95% CI)	Model 2 OR (95% CI)	Model 3 OR (95% CI)
Q1 (reference)	419	131	1	1	1
Q2	412	109	0.43 (0.31–0.67)*	0.37 (0.28–0.59)*	0.39 (0.28–0.61)*
Q3	415	81	0.32 (0.25–0.52)*	0.27 (0.23–0.51)*	0.25 (0.19–0.34)*
Q4	413	61	0.23 (0.19–0.27)*	0.16 (0.11–0.22)*	0.13 (0.08–0.17)*

Model 1: unadjusted; Model 2: adjusted for sex and grade; Model 3: adjusted for sex, grade, exercise duration, exercise frequency, and exercise period 

*P < 0.05 indicates statistical significance

Stratified logistic regression analyses by sex and grade demonstrated that the inverse relationship between physical fitness and injury risk was consistent across subgroups (Table 3). Both male and female students with higher fitness scores had significantly lower odds of sports injury compared to those in the lowest fitness quartile (Q1). For example, among male students, the adjusted OR for sports injury was 0.36 in Q4 and 0.21 in Q3 (compared to Q1, both P < 0.05); among female students, the corresponding ORs were 0.32 in Q4 and 0.36 in Q3 (both P < 0.05). Similarly, when stratified by grade level, higher fitness was associated with reduced injury risk in each grade category. First-year students in the highest fitness quartile had an adjusted OR of 0.23 for injury compared to Q1, while second-year and third-year students in Q4 had ORs of 0.15 and 0.12, respectively (all P < 0.05 vs. Q1). Although the exact OR values varied somewhat between subgroups (ranging, for instance, from approximately 0.12 to 0.41 for Q4 vs. Q1 across different sex and grade strata), the direction of the association was uniformly protective. There was no significant interaction effect by sex or grade, suggesting that the negative association between physical fitness and sports injury risk did not differ appreciably between males and females or among the different grade levels.

Table 3.

Stratified logistic regression analysis of the association between physical fitness score quartiles and sports injury risk by sex and grade level (OR [95% CI])

Fitness Quartile	Male (n = 846)	Female (n = 813)	Grade 1 (n = 712)	Grade 2 (n = 499)	Grade 3 (n = 448)
Q2	0.27 (0.13–0.43)*	0.39 (0.24–0.57)*	0.41 (0.21–0.78)*	0.33 (0.24–0.68)*	0.26 (0.14–0.51)*
Q3	0.21 (0.09–0.38)*	0.36 (0.19–0.52)*	0.32 (0.20–0.51)*	0.27 (0.17–0.43)*	0.21 (0.09–0.46)*
Q4	0.36 (0.22–0.71)*	0.32 (0.20–0.66)*	0.23 (0.20–0.43)*	0.15 (0.14–0.36)*	0.12 (0.06–0.27)*

Odds ratios (ORs) were calculated with Q1 (lowest quartile) as the reference group. Grade 1 = Grade 10 and first-year vocational students; Grade 2 = Grade 11 and second-year vocational students; Grade 3 = Grade 12 and third-year vocational students

*P < 0.05 indicates statistical significance

Non-linear relationship between physical fitness score and injury risk

Figure 1 illustrates the relationship between continuous physical fitness scores and the predicted probability of sports injury after adjusting for potential confounders (grade, sex, and exercise duration/frequency/period). The smoothed curve indicates a non-linear association, with a noticeable inflection point around a fitness score of 65. As the physical fitness score increased up to approximately 65, the risk of sports injury declined markedly. In quantitative terms, a piecewise analysis revealed that for students with fitness scores below 65, each 1-point increase in the fitness score was associated with roughly a 14% reduction in the odds of sports injury (OR per point = 0.86, 95% CI: 0.83–0.89, P < 0.05). However, once the fitness score reached about 65 or above, further increases in the score were no longer associated with a statistically significant change in injury risk (OR per point = 0.98, 95% CI: 0.95–1.02, P > 0.05). This suggests the presence of a threshold effect: improvements in physical fitness are linked to substantially lower sports injury risk up to a certain fitness level, beyond which the protective effect plateaus and additional gains in fitness confer no significant extra reduction in injury risk.

Fig. 1.

Nonlinear association between physical fitness score (red dashed line) and probability of sports injury, with 95% confidence intervals (blue circles and lines). The red dashed line represents the smoothed relationship between physical fitness score and the predicted probability of sports injury, estimated using multivariate logistic regression adjusted for sex, grade level, exercise duration, frequency, and period. The blue circles and lines represent the 95% confidence intervals (CI)

Discussion

In this large cross-sectional study of Chinese high school students, we found an overall sports injury prevalence of 23.0%, with boys (26.7%) reporting a higher prevalence of injuries than girls (19.2%). This gender disparity is consistent with prior research showing that male adolescents generally sustain more sports injuries than females [17]. One plausible explanation is that boys may engage more frequently in physically intense activities and exhibit greater risk-taking behaviors during play, factors that have been identified as contributors to youth sports injuries [18]. Our findings align with international data; for example, a recent Canadian survey reported that approximately 45% of high school students experienced at least one sports-related injury in a year, with higher rates observed among males [19]. While methodological differences (e.g., injury definitions and exposure levels) complicate direct comparisons, the higher injury prevalence among male students in our sample reflects a broader global trend.

Notably, some sport-specific studies suggest that when boys and girls participate in the same sports under similar conditions, their injury rates can be comparable [20]. However, the overall lower injury prevalence among girls in our study may reflect differences in sport participation profiles as well as potential physiological factors, rather than differences in collision-sport exposure. In fact, Chinese school sports programs generally offer fewer collision sports (e.g., no American football) compared to Western settings, which may partly explain why our observed overall prevalence was somewhat lower than that reported in North America and Europe [21]. Prior literature also notes that female athletes can be more prone to specific injuries (such as knee ligament injuries) due to biomechanical and neuromuscular factors, but males still sustain a greater absolute number of injuries because of higher overall participation rates and potentially less risk-averse behavior [22]. In summary, our sex-specific findings reinforce known epidemiological patterns and highlight the need for targeted injury prevention strategies in both sexes, while acknowledging contextual differences in school sports participation.

From a public health perspective, the injury prevalence of 23% in Chinese high schoolers is substantial, though somewhat lower than reports from North America and Europe [21]. This difference may be attributable to variations in school sports culture and the relative absence of high-collision activities in Chinese programs, which often emphasize structured and controlled exercise. Nevertheless, our data indicate that nearly one in four adolescents here experience a sports injury, underlining a significant public health issue. These injuries can disrupt academic participation and discourage youth from further physical activity, creating a vicious cycle counteracting fitness promotion effort. Therefore, understanding and mitigating the causes of these injuries in both boys and girls is critical.

Our second major finding was a robust inverse association between physical fitness levels and sports injury risk. Students with higher composite fitness scores experienced significantly fewer injuries, a trend observed across both sexes and all grade levels. This suggests that fundamental physical fitness – including attributes like muscular strength, endurance, speed/agility, and flexibility as assessed by our standardized tests – plays a protective role in youth sports safety. These results align closely with the broader literature. For example, a cohort study of adolescent female soccer players found that those with higher aerobic capacity (VO₂max) in preseason had markedly lower odds of in-season injury, with each unit increase in fitness conferring reduced injury risk [23]. Similarly, evidence from military and athletic populations has long shown that poor conditioning is a strong predictor of injuries; recruits or athletes with low cardiorespiratory fitness and inadequate muscle strength are significantly more likely to suffer musculoskeletal injuries during training [24]. Our findings in a general adolescent student sample reinforce these observations—indicating that fitness is a key determinant of resilience against injury even in routine school sports and physical education.

Several plausible mechanisms explain why higher fitness protects against injury. Physiologically, well-conditioned students have stronger muscles, ligaments, and tendons that can better support joints and absorb the shocks or sudden forces that occur in sports [25]. Good neuromuscular fitness (balance, coordination, and muscle control) helps athletes maintain proper technique and joint alignment during dynamic movements, reducing the likelihood of falls, twists or awkward landings that cause injuries [26]. This is particularly relevant for injuries like ankle sprains or knee ligament tears—deficits in neuromuscular strength and coordination are known risk factors for such injuries, especially in adolescents who are still developing motor control [27]. In our study, students with higher fitness scores likely had better neuromuscular conditioning, contributing to more stable and controlled movement patterns that prevented acute injuries (e.g., sprains, strains) during sports. In fact, prior research has demonstrated that targeted neuromuscular training programs (including balance, strength, and agility exercises) can reduce sports injury rates in youth by as much as 30–50%. Such programs improve athletes’ muscle activation and reflexes, lending empirical support to the idea that fitter, stronger musculoskeletal systems are more injury-resistant [28]. Our findings echo this: students who achieved higher scores in strength-and power-oriented fitness tests (e.g., standing long jump, grip strength) were less prone to injury, underlining the value of muscular fitness in injury prevention.

Cardiorespiratory fitness is another critical component. Students with superior endurance capacity can sustain activity with less fatigue, and fatigue is a well-documented risk factor for injury [29]. As players tire, their form deteriorates and reaction time slows, making injuries more likely. While we did not collect contextual details such as the timing or mechanism of injury, it is plausible—based on prior literature—that greater endurance capacity helps mitigate fatigue-related declines in movement control, which are linked to higher injury risk. Our findings therefore suggest an association rather than a direct causal explanation. Better fitness may also indicate more frequent engagement in physical activity, which could confer technical experience and adaptive changes (e.g., stronger bones and connective tissues) that protect against injury. Moreover, fit students might recover more quickly from minor stresses and have fewer chronic deficits (like poor flexibility or weak core muscles) that predispose to injury. These interpretations remain speculative and should be validated in future studies that incorporate detailed injury context data.

Perhaps the most novel finding in our study was the non-linear relationship between fitness scores and injury risk. We observed a clear inflection point around a fitness score of 65 (on our composite 0–100 scale). Below this threshold, each 1-point increase in the fitness score was associated with a substantial reduction in injury risk—in other words, improving fitness from very low levels yielded large safety benefits. However, once students attained a fitness score of 65 or above (approximately equivalent to a “moderate” fitness level), the protective effect plateaued. Additional gains in fitness above this level conferred little extra reduction in injury risk; injury rates in the high-fitness group were relatively flat compared to the moderate-fitness group. This plateau suggests diminishing returns to fitness in terms of injury prevention: having very poor fitness is a major liability, but once a reasonable fitness level is achieved, other factors may become the limiting determinants of injury risk. This kind of threshold phenomenon has been hinted at in prior research. For instance, functional movement screenings of college athletes in China found that injury risk was markedly elevated below a certain composite movement score (FMS score < 17.5), whereas athletes above that cut-off had comparatively lower and more uniform injury rates [14]. Such findings imply that a baseline level of physical competency is necessary to safely participate in sports—below that level, injury risk rises sharply. In our study, the threshold of 65 was not a predefined cutoff but rather emerged empirically from piecewise regression analysis, where the decline in injury risk began to plateau. Moreover, this value corresponds approximately to the “qualified” category of the 2014 National Student Physical Health Standards, in which scores of 60–70 denote the minimum passing threshold for Chinese adolescents. Thus, in this context, a score of 65 should be interpreted as a relative indicator of moderate or adequate fitness within this population, rather than as an absolute or universal benchmark. Each incremental improvement for them (e.g., improving strength, flexibility, or endurance) sharply lowers risk, likely by rectifying specific weaknesses (such as poor core stability or lack of stamina) that would otherwise lead to injury. This is encouraging, as it means that interventions aimed at the least-fit students could yield disproportionately large reductions in injury incidence. Even simple measures—for example, targeted conditioning for students who fail to meet fitness benchmarks—might substantially improve their safety in PE and sports.

Above the threshold of a physical fitness score of 65, the plateau suggests that once moderate fitness is achieved, further fitness improvements alone may not translate into additional injury protection. Although our study did not examine the specific circumstances of injuries (e.g., contact vs. non-contact, playing surface, or sport type), prior research indicates that extrinsic factors may become more influential at higher fitness levels [30]. Therefore, while highly fit students in our sample still experienced injuries, these events are likely attributable to external exposures not captured in our dataset, such as training volume or sport-specific risks. This aligns with the concept of a U-shaped relationship between training volume and injury: very low activity (with poor fitness) is risky, but very high activity can also elevate injury risk due to overuse or insufficient recovery [31]. A recent study in 15–16 year-olds supports this dual risk pattern—it identified two distinct groups prone to PE injuries: (a) students with very low activity/fitness (“fragile inactive” youth unfamiliar with exercise) and (b) students with very high activity (sport specialists with heavy training, prone to repeat injuries) [25]. The latter group likely corresponds to those above our fitness threshold: they are fit, but their high sport exposure can lead to recurrent injuries (especially if previous injuries aren’t fully rehabilitated). A parallel can be drawn with our findings—improving fitness greatly helps those below the threshold, but for those above it, injury prevention needs to focus on other factors like adequate rest, periodized training, proper technique, and injury management.

Our finding of a non-linear fitness–injury relationship also carries practical significance for schools. It implies there may be a “safe fitness zone” that students should be helped to achieve. Fitness score 65 (in our battery) might be viewed as a target minimum for all students. Those falling below this level could be prioritized for remedial physical training and closer supervision during sports. Conversely, for those already above the threshold, additional fitness gains might yield smaller safety benefits – so coaches should ensure these students focus on sport-specific skill technique and recovery practices to prevent injuries, rather than just on accumulating more fitness. Overall, recognizing this threshold effect allows for a more tailored approach: raise the fitness of the most unfit (to drastically cut their injury risk), and manage the training of the fittest (to avoid overuse and burnout). Future longitudinal research could explore if there is indeed a causal cutoff point and whether interventions that bring students over that bar lead to demonstrable reductions in injury incidence.

This study’s strengths include a large, representative sample of high school students with objectively measured fitness and self-reported injury data, allowing us to directly examine associations between physical fitness levels and injury outcomes. We employed standardized national physical fitness tests, enhancing the reliability and comparability of our fitness measures. Moreover, by capturing students from all grades and both sexes in urban Shanghai, our findings likely reflect broader adolescent trends (at least in similar urban settings). However, several limitations warrant consideration. First, the data are region-specific (Shanghai city) and may not generalize to all Chinese adolescents, especially those in rural areas or different sports environments. Second, sports injuries were reported via questionnaire, introducing potential recall bias or misclassification (minor injuries might be forgotten, or the definition of “injury” might vary between respondents). To minimize this risk, we followed the standardized administration protocol of a validated instrument, in which participants were provided with the established definition of sports injury during survey completion [15]. This clarification ensured consistency with the original questionnaire design, although some underreporting or overreporting remains possible. Third, the cross-sectional design limits causal inference—while we observed that low fitness was correlated with higher injury risk, we cannot definitively conclude that poor fitness caused injuries. It is conceivable, for example, that students who had injuries subsequently exercised less and thus scored lower on fitness tests. A bidirectional relationship may exist, and longitudinal studies are needed to confirm whether improvements in fitness lead to reductions in injury incidence. Finally, we did not account for all potential confounders such as socioeconomic status, nutrition, or detailed training hours, which might influence both fitness and injury risk. Despite these limitations, our study provides valuable insights into the associations between fitness and injuries in a youth population and highlights both the promise of fitness improvement as an injury countermeasure and the need for holistic injury prevention strategies in high school sports.

Conclusion

In this large cross-sectional study of Chinese high school students, we observed a substantial sports injury prevalence of 23.0%, significantly higher in boys than girls. Physical fitness was inversely associated with injury risk, displaying a clear nonlinear threshold effect: injury risk markedly decreased with improved fitness scores up to a threshold of 65, beyond which additional fitness improvements yielded minimal protective benefits. These findings highlight the critical importance of baseline fitness enhancement as a targeted injury-prevention strategy for adolescents. Schools should prioritize interventions aimed at elevating fitness among lower-performing students while emphasizing comprehensive approaches—including load management, neuromuscular training, and safe sports practices—for students already achieving moderate-to-high fitness levels. Future longitudinal studies are warranted to verify causality and refine preventive interventions.

Abbreviations

BMI

Body Mass Index

CI

Confidence Interval

FMS

Functional Movement Screen

OR

Odds Ratio

PE

Physical Education

PF

Physical Fitness

Q1-Q4

Quartiles 1 to 4 (for fitness score distribution)

VO₂max

Maximal Oxygen Uptake

Authors’ contributions

QJZ contributed substantially to the study’s conception, design, and manuscript drafting. SZ and SLZ were responsible for data collection, entry, and statistical analysis. LZ provided critical revision of the manuscript for important intellectual content and supervised the entire research process. All authors read and approved the final version of the manuscript.

Funding

Not applicable.

Data availability

Data and materials of this project are available in the repository of the Shanghai Qunxing Vocational and Technical School. The specific datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by Ethics Committee of Shanghai Qunxing Vocational and Technical School (No. #202004065). All methods were carried out in accordance with relevant guidelines and regulations. This study was carried out in compliance with the Declaration of Helsinki, the Ethical approvals are detailed in the Supplementary. Parents/guardians of the students participating in the study were provided with informed consent forms, and their consent was obtained for their children's participation. Prior to their participation, all students were informed in a manner consistent with the information provided to their parents/guardians, and their individual consent was also obtained.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.