Impact of a school based flag football intervention on fitness outcomes in Chinese children aged 9 to 11 years: a randomized controlled trial

Abstract

Purpose

To evaluate the effects of a 10-week school-based flag football programme on fitness outcomes in Chinese children aged 9 to 11 years.

Methods

A two-arm, parallel-group randomized controlled trial was conducted in a public primary school in Shanghai, China. Children (n = 147) were randomized to either flag football (n = 74; mean age, 9.7 ± 0.5 years; 30 girls) or usual physical education (PE; n = 73; mean age, 9.7 ± 0.5 years; 32 girls). The intervention comprised 60-min sessions, delivered twice weekly for 10 weeks. Outcomes were assessed at baseline and post-intervention: the 20 m shuttle run test (laps), estimated maximal oxygen uptake (VO₂max), the 1-min jump rope test (turns), and the 50 m dash (seconds). Primary analyses followed the intention-to-treat principle using multiple imputation and repeated-measures models; subgroup analyses were conducted by sex and baseline weight status. Session intensity was monitored through heart rate (HR) in a subsample.

Results

Post-intervention data were available for 128 children (87.1%). Compared with usual physical education, the flag football group showed greater improvements in 1-min jump rope performance (between-group difference in change = 7.4 turns; p = 0.03) and 50 m dash time (between-group difference in change = -0.3 s; p = 0.013). No between-group differences in change were observed for 20 m shuttle run performance (0.7 laps; p = 0.32) or estimated VO₂max (0.19 mL·kg⁻¹·min⁻¹; p = 0.37). Subgroup analyses revealed the most significant effects in boys with normal weight for estimated VO₂max and 20 m shuttle run, boys across BMI status groups for sprint performance, and girls across BMI status groups for jump rope.

Conclusions

A PE-embedded flag football programme effectively improved sprint speed and 1-min jump rope performance. The observed sex- and weight-specific responses suggest the need for differentiated task progression in school-based interventions to maximize fitness benefits.

Trial registration

Trial registration: ISRCTN registry (Current Controlled Trials), ISRCTN62093106; registered 04 February 2026 (retrospectively registered).

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28081-x.

Introduction

Insufficient physical activity and declining physical fitness in school aged children have emerged as major public health concerns, with implications for cardiometabolic health across the life course [1]. Exercise based interventions can produce meaningful health benefits and promote sustained engagement in physical activity, yet effects vary by population, setting, and the feasibility of programme delivery at scale [2]. School based strategies are particularly important because they can reach children regardless of household resources and can be embedded within routine physical education; however, many programmes are constrained by limited instructional time, competing curricular demands, and the need for approaches that are safe, engaging, and easy to implement [3].

Sport based interventions represent a pragmatic pathway to increase moderate to vigorous activity during the school day while concurrently developing movement competence. Team games may be especially attractive to children because they are social, goal directed, and intrinsically motivating [4]. Within this context, flag football is a non-contact, invasion style sport that emphasizes repeated bouts of running, acceleration, deceleration, rapid changes of direction, and reactive decision making, alongside throwing and catching skills. These demands align with fitness components that are both clinically meaningful and measurable in school settings, including cardiorespiratory fitness, speed, and muscular endurance [5, 6]. In addition, like other scalable movement programmes that require minimal space and no specialized equipment, flag football can be delivered with low-cost materials and standardized lesson plans, supporting dissemination in resource-variable school environments [7].

Despite growing interest in using sport as a vehicle for health promotion, rigorous evidence on flag football as a structured school-based intervention remains limited, particularly in mainland China. Most existing school sport programmes have evaluated broad multicomponent curricula or traditional activity modules, making it difficult to determine whether a specific sport can deliver meaningful fitness benefits within routine physical education [8]. Evaluating an individual sport is important because implementation feasibility, movement demands, safety profile, and training effects may differ across activities. Moreover, interventions that depend on equipment intensive training or require heightened safety monitoring may face barriers to adoption and sustainability in routine practice [9]. Flag football is of particular interest because it is non-contact, low-cost, and easily standardized for school delivery, while targeting movement patterns relevant to key fitness outcomes. There is therefore a need for well controlled trials that test feasible, curriculum compatible sport programmes and quantify effects on standardized fitness outcomes. Child level heterogeneity is also likely to influence responsiveness to sport-based training. Sex related differences in baseline fitness, movement patterns, and participation preferences may shape training adaptations, and adiposity may modify both physiological response and perceived exertion [10]. Consequently, interventions that appear modest at the group level may still yield clinically relevant benefits within specific subgroups. Trials that incorporate prespecified subgroup analyses can help identify which children benefit most and inform more targeted implementation strategies.

To address these gaps, the present randomized controlled trial examined whether a 10-week, school-based flag football programme embedded within routine physical education could improve key fitness outcomes in Chinese children aged 9 to 11 years. This study adds to the literature by evaluating a non-contact, low-cost, and curriculum-compatible sport programme in a real-world primary school setting in mainland China, with prespecified analyses exploring whether intervention responses differed by sex and baseline BMI status.

Methods

Study design and setting

This study used a school-based, two-arm, parallel-group randomized controlled design to evaluate the effects of a 10-week flag football intervention on physical fitness in primary school children. The trial was conducted in a public primary school in Shanghai, China, during regular school hours. Trial conduct and reporting followed the CONSORT guidelines for randomized controlled trials (Supplemental Digital Content 1).

Participants

Children were eligible if they were enrolled in grades 3 to 5, were between 9 and 11 years of age, and had no medical contraindication to moderate-to-vigorous physical activity. Exclusion criteria included physician-diagnosed chronic disease, musculoskeletal or neurological disorders that limited participation in running or jumping activities, and any other condition judged by the school physician or research team to make participation unsafe.

From six classes, 180 children were invited to participate. Of these, 147 children and their parents provided written informed consent and completed baseline assessments. Before recruitment was completed, an a priori sample size calculation was performed using G*Power. Based on a two-sided test for the difference between two independent means, with an alpha level of 0.05, power of 0.80, an allocation ratio of 1, and a medium effect size (Cohen’s d = 0.5), the required sample size was estimated at 128 participants, or 64 per group. Allowing for an anticipated attrition rate of 10%, the target recruitment was increased to 144 participants. Participants were then randomized to the intervention group (n = 74) or the control group (n = 73). Complete pre- and post-intervention fitness data were obtained from 128 children (intervention group, n = 65; control group, n = 63), who comprised the complete-case sample, and all randomized participants were included in the intention-to-treat analyses (Fig. 1).

Fig. 1.

Participant flow through the trial

Prior to randomization, height and body mass were measured barefoot in light clothing, and body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Children were classified using age- and sex-specific BMI cut points recommended by the Chinese Working Group on Obesity in Children [11]. Normal weight was defined as BMI below the overweight threshold, and overweight or obesity was defined according to the corresponding reference cut points for each age and sex. No participants were classified as underweight.

The study protocol was approved by the Ethics Committee of Shanghai University of Sport. All participating children and their parents or legal guardians were informed of the study aims, procedures, potential risks, and benefits, and provided written informed consent before enrolment.

Randomization

After completion of baseline assessments, children were randomly allocated in a 1 to 1 ratio to the flag football intervention group or the control group at the individual level. The allocation sequence was generated by an independent researcher using a computer-generated random number list with variable block sizes within each class and was implemented with sequentially numbered, opaque, sealed envelopes. Randomization was conducted within classes to minimize disruption to teaching. Given the nature of the intervention, participants and instructors could not be blinded to group assignment. Outcome assessments were conducted by trained postgraduate students who were independent of intervention delivery and were not informed of allocation whenever feasible.

Interventions

Flag football intervention

The intervention group participated in a structured flag football program for 10 consecutive weeks. Two sessions per week were delivered on Monday and Friday afternoons during regular physical education periods. Each session lasted 60 min and followed a standardized lesson plan comprising three components: (a) a 10 min warm up with dynamic stretching, jogging, and movement preparation activities; (b) a 40 min main segment with skill drills and small-sided games designed to progressively develop flag football techniques and game understanding, including familiarization with the ball, throwing and catching, position specific movement patterns, offensive and defensive alignment, one versus one defending, and modified games emphasizing running, cutting, and change of direction; and (c) a 10 min cool down with low intensity activities and static stretching of the major muscle groups of the lower and upper limbs. This schedule was determined by the school timetable and provided 120 min of intervention exposure per week.

Exercise intensity during each session was monitored in a sub-sample of 12 children per class using Actiheart heart rate monitors (Cambridge, United Kingdom) [12]. Heart rate monitoring was conducted only in the intervention group to characterize the physiological intensity of the flag football sessions. To improve representativeness, monitored participants were selected within each class to achieve, where possible, balanced representation by sex and baseline BMI status. The same children were monitored across sessions whenever feasible; if a selected child was absent or a recording was incomplete, a replacement from the same class was used. Before each session, individual characteristics were entered into the device software, and monitors were attached to the chest with two disposable electrodes. Heart rate was sampled every 30 s. Target intensity for the intervention followed American College of Sports Medicine recommendations, with exercise prescribed at 60 to 90 percent of age predicted maximal heart rate (HRmax), where HRmax was estimated as 208 minus 0.7 multiplied by age in years [13]. Time spent below 64%, between 65 and 76%, and above 77% of HRmax was used to characterize light, moderate, and vigorous intensity, respectively [14].

Control condition

Children allocated to the control group continued to receive the school’s usual physical education curriculum and did not participate in any flag football activities during the study period. Usual physical education typically consisted of three 35-min sessions per week that included a variety of calisthenics and ball games but no systematic training in flag football techniques. No additional organized exercise program was introduced by the research team in the control group. Although the control group had three sessions per week, total weekly instructional time was broadly comparable to that of the intervention group (105 vs 120 min per week). This approach was chosen to preserve routine school practice and evaluate the feasibility of flag football under real-world PE conditions.

Outcome measures

Outcomes were prespecified as primary and secondary measures of health-related physical fitness. The primary outcomes were performance on the 20-m shuttle run test, estimated maximal oxygen uptake, one minute jump rope test, and 50-m dash. These outcomes were selected to capture cardiorespiratory fitness, task-specific jump rope performance, and speed. Secondary outcomes included upper body strength (handgrip strength), lower limb power (vertical jump), trunk muscular endurance (one minute sit ups), flexibility (sit and reach), and anthropometric measures (height, weight, and BMI).

All physical fitness assessments were conducted within one week before the start of the intervention and within one week after completion of the 10-week program. Testing followed the National Student Physical Fitness Standard protocols and was administered in the school gymnasium or playground by trained graduate students who underwent centralized instruction prior to data collection [15]. Participants completed a standardized warm up before testing and were encouraged verbally during each test but were not given feedback on interim performance.

Cardiorespiratory fitness

Cardiorespiratory fitness was assessed using the 20-m shuttle run test. After warm up, children ran back and forth between two lines 20 m apart, synchronized to audio signals from a calibrated compact disc. The initial speed was 8.5 kms per hour and increased by 0.5 kms per hour every minute. The test was terminated when a child failed twice consecutively to reach the line before the audio signal. The number of completed shuttles (laps) was recorded as the raw score. Final running speed was calculated as 8 plus 0.5 times the last stage completed, and VO2max (mL·kg − 1·min − 1) was estimated using the Leger prediction Eq. [16].

Jump rope performance

Task-specific jump rope performance was assessed using a one-minute jump rope test. Children practiced the correct technique and then performed continuous two-foot jump rope for 60 s. The total number of correctly completed jumps was counted and expressed as turns per minute. This outcome was interpreted as a functional performance measure influenced by repeated jumping ability, coordination, rhythm, neuromuscular control, and cardiorespiratory demand, rather than as an isolated measure of lower-limb muscular endurance.

Speed

Speed was assessed using a 50-m sprint. Participants started from a standing position with the front foot placed exactly 0.5 m behind the start line and were instructed to run as fast as possible to the finish line after the start signal. Time to cover 50 m was measured using handheld electronic stopwatches and reported to the nearest 0.1 s to better reflect the practical precision of manual timing. Two trials were performed with a 3-min rest interval between attempts, and the fastest time was used for analysis.

BMI subgroup classification

For predefined subgroup analyses, children were classified at baseline as normal weight or overweight or obesity based on age- and sex-specific BMI cut points proposed by the Chinese Working Group on Obesity in Children [11]. This classification has been widely used in national surveillance studies and intervention research in Chinese schoolchildren.

Statistical analysis

All analyses were conducted in SPSS version 22.0 (IBM Corp). Continuous variables are reported as mean and standard deviation. Baseline characteristics were compared between groups using independent samples t tests for continuous variables and chi-square tests for categorical variables.

Primary analyses were performed according to the intention-to-treat principle, including all 147 randomized participants. Missing post-intervention outcomes were imputed using multiple imputation by chained equations under a missing at random assumption. The imputation model included group allocation, BMI category, age, sex, and baseline values of each outcome. Ten imputed data sets were created, analyzed separately, and combined using Rubin’s rules.

Intervention effects were evaluated on the imputed data using repeated measures analysis of variance, with time (baseline and post intervention) as the within participant factor and group (intervention and control) and BMI category (normal weight and overweight or obesity) as between participant factors. The primary effect of interest was the group by time interaction. Between-group differences in change are reported with 95% confidence intervals to indicate the precision of the estimated intervention effects. Effect modification by BMI category was assessed by the time by group by BMI category interaction. When interaction terms were significant, simple effects were examined within each group and BMI stratum with Bonferroni adjustment.

Sensitivity analyses were conducted in complete cases with both baseline and post intervention measurements (n = 128; intervention group, n = 65; control group, n = 63) using the same model specification. All tests were two-sided with a significance level of 0.05. Effect sizes are presented as partial ηp2 and interpreted as small (0.01 to less than 0.06), medium (0.06 to less than 0.14), and large (0.14 or greater).

Results

Participants and follow up

Of 180 children invited, 147 provided consent and were randomized (flag football, n = 74; control, n = 73; Fig. 1). The intention to treat population comprised all 147 randomized participants. Complete pre and postintervention fitness data were available for 128 children (flag football, n = 65; control, n = 63), yielding an overall follow up rate of 87.1%.

Baseline characteristics

Baseline characteristics of participants with complete outcome data are presented in Table 1. Mean (SD) age was 9.7 (0.8) years; 62 of 128 participants (48.4%) were girls; and 57 of 128 (44.5%) were classified as having overweight or obesity. Baseline daily moderate to vigorous physical activity (MVPA) was 46.2 (18.4) min/day in the intervention group and 41.9 (17.0) min/day in the control group. No statistically significant between-group differences were observed at baseline (all p > 0.05).

Table 1.

Baseline characteristics of the study participants

Characteristic	Intervention (n = 65)	Control (n = 63)	Total (N = 128)
Age, y, mean (SD)	9.7 (0.5)	9.7 (0.5)	9.7 (0.5)
Female sex, No. (%)	30 (46.2)	32 (50.8)	62 (48.4)
Height, cm, mean (SD)	144.0 (5.6)	143.8 (6.7)	143.9 (6.2)
Weight, kg, mean (SD)	39.9 (8.5)	40.9 (8.5)	40.4 (8.5)
Body mass index, kg/m2, mean (SD)	19.1 (3.3)	19.6 (3.1)	19.4 (3.2)
Overweight or obesity, No. (%)	29 (44.6)	28 (44.4)	57 (44.5)
Daily MVPA, min/day, mean (SD)	46.2 (13.6)	41.9 (13.2)	44.0 (13.5)

Primary outcomes

Primary outcomes at baseline and postintervention, and between group differences in change based on ITT analyses, are shown in Table 2. Compared with usual physical education, the flag football intervention resulted in a greater improvement in 1 min jump rope performance (between-group difference in change, 7.4 repetitions; 95% CI, 0.7 to 14.1; p = 0.03) and a greater reduction in 50 m sprint time (between-group difference in change, −0.3 s; 95% CI, −0.5 to −0.1; p = 0.013). No evidence of a between group difference in change was observed for 20 m shuttle run performance (0.7 laps; 95% CI, −0.7 to 2.1; p = 0.32) or estimated VO₂max (0.19 mL·kg⁻¹·min⁻¹; 95% CI, −0.23 to 0.61; p = 0.37).

Table 2.

Primary and secondary outcomes at baseline and post-intervention in the intervention and control groups (ITT Analysis)

Measure	Intervention (n = 65)		Control (n = 63)		Between-group difference in change (95% CI)	P value
	Baseline	Post	Baseline	Post
20 m shuttle run, laps	26.7 ± 9.0	32.2 ± 9.5*	22.8 ± 7.5	27.7 ± 8.8*	0.7 (−0.7 to 2.1)	0.32
Estimated VO2max, ml·kg⁻1·min⁻1	46.24 ± 2.94	47.64 ± 3.07	45.15 ± 2.63	46.37 ± 2.69	0.19 (−0.23 to 0.61)	0.37
1-min rope skipping, repetitions	107.3 ± 27.7	117.9 ± 31.4*#	113.3 ± 28.8	116.4 ± 30.3*	7.4 (0.7 to 14.1)	0.03
50 m sprint, s	10.1 ± 0.9	9.7 ± 0.7*#	9.9 ± 0.8	9.8 ± 0.9	−0.3 (−0.5 to −0.1)	0.01

^*p < 0.05 vs preintervention

^#p < 0.05 vs control group

P < 0.05 vs preintervention

#P < 0.05 vs control group

Bold values indicate statistically significant between-group differences (P < 0.05)

Sensitivity analyses restricted to participants with complete data yielded similar results (Supplemental Digital Content 2, Table S2.1). In complete-case analyses, improvements remained greater in the intervention group for jump rope performance (increase from 108.1 ± 27.2 to 118.8 ± 30.9 repetitions vs 113.5 ± 28.5 to 116.2 ± 30.1 in controls) and 50 m sprint time (decrease from 10.1 ± 0.9 to 9.6 ± 0.7 s vs 9.9 ± 0.8 to 9.8 ± 0.9 s in controls), while between group differences for 20 m shuttle run and estimated VO₂max remained non-significant. These findings support the robustness of the ITT results.

Stratified findings by sex and BMI category

In sex and BMI stratified analyses, improvements in cardiorespiratory fitness were most evident among boys with normal weight. In this subgroup, the between-group difference in change was 5.9 laps for the 20 m shuttle run (95% CI, 0.1 to 11.8; p = 0.048) and 1.9 mL·kg⁻¹·min⁻¹ for estimated VO₂max (95% CI, 0.17 to 3.63; p = 0.032). For sprint performance, improvements were observed among boys with normal weight (difference in change, −0.4 s; 95% CI, −0.75 to −0.05; p = 0.026) and boys with overweight or obesity (difference in change, −0.5 s; 95% CI, −0.98 to −0.02; p = 0.040). For jump rope performance, improvements were observed in girls with normal weight (difference in change, 13.5 turns·min⁻¹; 95% CI, 1.1 to 25.8; p = 0.033) and girls with overweight or obesity (difference in change, 10.8 turns·min⁻¹; 95% CI, 3.1 to 18.5; p = 0.008) (Table 3).

Table 3.

Subgroup analyses of primary and secondary outcomes by sex and baseline weight status (ITT Analysis)

Sex	Baseline weight status	Measure	Intervention		Control		Difference in change (95% CI)	P value
			Baseline	Post	Baseline	Post
Boys	Normal weight (n = 18; control n = 17)	20-m shuttle run, laps	29.27 ± 8.60	39.20 ± 5.20*	27.20 ± 6.11	31.20 ± 6.96*	5.9 (0.1 to 11.8)	0.048
		Estimated VO₂max, mL·kg⁻1·min⁻1	46.87 ± 3.11	49.74 ± 1.99*	46.34 ± 2.77	47.31 ± 2.29#	1.9 (0.17 to 3.63)	0.032
		Jump rope test (turns·min⁻1)	106.25 ± 25.97	114.69 ± 20.13	113.27 ± 25.23	116.13 ± 20.25	5.6 (−8.8 to 20.0)	0.437
		50-m dash (s)	9.7 ± 0.5	9.4 ± 0.5*	9.4 ± 0.6	9.5 ± 0.8	−0.4 (−0.75 to −0.05)	0.026
Boys	Overweight/obesity (n = 22; control n = 19)	20-m shuttle run, laps	23.47 ± 9.11	28.58 ± 9.11*	18.13 ± 8.72	23.00 ± 8.54*	0.2 (−4.2 to 4.7)	0.914
		Estimated VO₂max, mL·kg⁻1·min⁻1	45.59 ± 3.26	46.85 ± 3.09*	43.53 ± 2.95	44.70 ± 3.23*#	0.09 (−1.19 to 1.37)	0.888
		Jump rope test (turns·min⁻1)	87.47 ± 25.40	95.37 ± 27.39	103.25 ± 27.07	111.19 ± 30.53	−0.04 (−12.9 to 12.8)	0.995
		50-m dash (s)	10.3 ± 0.9	9.9 ± 0.6*	10.1 ± 0.8	10.2 ± 0.9	−0.5 (−0.98 to −0.02)	0.040
Girls	Normal weight (n = 23; control n = 23)	20-m shuttle run, laps	30.15 ± 8.98	33.40 ± 9.46*	24.50 ± 6.84#	29.90 ± 9.86*	−2.2 (−6.4 to 2.1)	0.317
		Estimated VO₂max, mL·kg⁻1·min⁻1	46.98 ± 2.78	47.69 ± 3.18	45.88 ± 1.77	47.21 ± 2.22*	−0.62 (−2.01 to 0.77)	0.373
		Jump rope test (turns·min⁻1)	124.00 ± 23.96	137.21 ± 27.04*	119.00 ± 32.85	118.75 ± 37.58	13.5 (1.1 to 25.8)	0.033
		50-m dash (s)	9.9 ± 0.8	9.4 ± 0.6*	9.9 ± 1.0	9.6 ± 0.9	−0.2 (−0.58 to 0.18)	0.290
Girls	Overweight/obesity (n = 11; control n = 14)	20-m shuttle run, laps	22.00 ± 5.33	25.70 ± 8.69	20.75 ± 4.18	25.75 ± 6.88	−1.3 (−5.9 to 3.3)	0.568
		Estimated VO₂max, mL·kg⁻1·min⁻1	44.97 ± 1.80	45.69 ± 2.58	44.59 ± 2.23	46.00 ± 2.14	−0.69 (−2.42 to 1.04)	0.418
		Jump rope test (turns·min⁻1)	113.30 ± 18.79	127.10 ± 36.00	117.00 ± 28.08	120.00 ± 29.66	10.8 (3.1 to 18.5)	0.008
		50-m dash (s)	10.5 ± 1.2	10.1 ± 0.7*	10.1 ± 0.9	9.9 ± 0.7	−0.2 (−0.81 to 0.41)	0.501

^#p < 0.05 vs control group

^*p < 0.05 vs preintervention

Complete-case sensitivity analyses (n = 128) showed a similar pattern of subgroup findings (Supplemental Digital Content 2, Table S2.2), with cardiorespiratory fitness gains most apparent among boys with normal weight, sprint improvements evident among boys across BMI categories, and jump rope improvements observed among girls regardless of baseline BMI category.

Exercise intensity monitoring

Heart rate monitoring during the flag football sessions indicated that the programme was delivered at a moderate to vigorous intensity overall. Mean (SD) heart rate was 144.2 (8.7) beats/min for boys with normal weight and 139.0 (7.2) beats/min for girls with normal weight, corresponding to 71.6% (4.0%) and 69.1% (3.5%) of predicted HRmax, respectively. Among participants with overweight or obesity, mean (SD) heart rate was 146.2 (7.6) beats/min for boys and 144.1 (8.2) beats/min for girls, corresponding to 72.5% (3.7%) and 71.6% (3.9%) of predicted HRmax, respectively. Across all monitored participants, mean (SD) heart rate was 143.2 (6.8) beats/min, equivalent to 71.1% (3.4%) of predicted HRmax.

Discussion

In this school based, two arm randomized controlled trial embedded within routine physical education, a 10-week flag football programme was implemented in Chinese children aged 9 to 11 years to evaluate its effects on key fitness outcomes. Overall, participation in flag football led to clearer improvements in 1-min jump rope performance and sprint performance than usual physical education, whereas group level changes in cardiorespiratory fitness were not clearly differentiated. Accordingly, the jump rope outcome should be interpreted as a task-specific functional performance measure rather than a direct or isolated indicator of lower-limb muscular endurance. Prespecified subgroup analyses indicated that responses were not uniform, with cardiorespiratory fitness gains more apparent among boys with normal weight, jump rope performance improving more consistently among girls across weight categories, and sprint improvements most evident among boys regardless of baseline weight status. These findings suggest that flag football is a curriculum compatible, game-based strategy that can enhance selected components of fitness in primary school children, while also highlighting the importance of considering sex and weight status when designing and scaling school sport interventions.

The preferential improvements in sprint speed and one minute jump rope performance are consistent with the specific movement profile of flag football and with prior school-based sport interventions in which neuromuscular and skill related outcomes tend to respond more readily than aerobic indices over relatively short programmes [17, 18]. Flag football sessions typically involve repeated short bursts of high intensity effort interspersed with brief recovery, with frequent accelerations, decelerations, and rapid changes of direction during small sided play. Such intermittent loading provides a training stimulus that is well suited to improving short distance running performance through enhanced force application, braking control, and efficient transitions between acceleration and reacceleration, rather than through sustained steady state work [19]. Improvements in jump rope performance are also biomechanically plausible because the test reflects rhythmic lower limb stiffness, stretch shortening cycle efficiency, and whole-body coordination. The repeated exposure to footwork, cutting patterns, and reactive movement demands in game contexts may strengthen neuromuscular control and movement timing, which could transfer to the rapid, continuous cyclical actions required for rope skipping [20].

From a health and educational perspective, gains in speed and lower limb muscular endurance align with fitness attributes that are highly visible in everyday play and common school-based fitness assessments, and may therefore be more readily perceived by children and teachers [21]. This salience has practical value for physical education because early, noticeable improvements may support enjoyment and continued participation, reinforcing engagement with movement opportunities within and beyond the school day.

Although aerobic fitness tended to improve over time in both groups, the intervention did not produce a clear additional benefit for 20 m shuttle run performance or estimated VO₂max at the whole sample level. This pattern aligns with school-based sport trials in which short duration or low dose programmes often yield more robust changes in neuromuscular outcomes than in cardiorespiratory fitness, particularly when the comparison group continues to receive regular physical education [9, 22]. A plausible explanation is that the incremental training stimulus delivered by two additional sessions per week over 10 weeks may have been modest relative to children’s usual activity opportunities, thereby reducing the contrast in total exposure between groups [22]. In addition, the monitored sessions were characterized by a mean heart rate around 70% of predicted HRmax, indicating an overall moderate physiological load rather than sustained high intensity exercise. The intermittent nature of game play may still have produced brief vigorous bursts, but mean session heart rate alone does not indicate how much time was spent at higher intensity thresholds within sessions. Finally, developmental heterogeneity in baseline fitness and maturation may have diluted group level effects over the relatively short intervention period [23]. These findings should therefore be interpreted cautiously, and future studies should quantify time spent above prespecified heart rate thresholds to better characterize the dose of vigorous activity achieved during flag football sessions.

Building on the overall effects, the prespecified subgroup analyses suggest that intervention responses varied by sex and baseline weight status, which is consistent with evidence that children’s adaptation to school based physical activity is heterogeneous [24]. A clear pattern emerged in which improvements in cardiorespiratory fitness outcomes were more apparent among boys with normal weight, sprint performance gains were most evident among boys across weight strata, and rope skipping performance improved more consistently among girls regardless of weight status. Similar effect modification has been reported in school interventions and sport-based programmes, where sex differences in baseline fitness, movement competence, and engagement with competitive or pursuit-oriented activities can influence the magnitude and type of training response [25]. Boys may accrue greater exposure to high velocity running and repeated accelerations during invasion games because of play style and task selection, while girls may benefit more from the repeated rhythm and footwork elements that support coordinated cyclic actions relevant to rope skipping [26]. Weight status may further shape responsiveness through mechanical and physiological constraints. Children with overweight or obesity experience higher relative loading during acceleration and deceleration, which may limit sustained high intensity running and therefore blunt improvements in shuttle based aerobic performance, yet intermittent tasks that emphasize short bursts of speed or neuromuscular efficiency may still be adaptable within this group [27, 28].

These subgroup findings should be interpreted cautiously because stratum specific sample sizes were modest, and the analyses are best viewed as hypothesis generating; confirmation in larger trials, ideally using cluster randomization and stratified sampling, is warranted. From an implementation perspective, the results support consideration of differentiated instruction and tiered task design within physical education, so that activity demands and progression can be optimized for sex and weight related differences in capability and response.

To support interpretation of the outcome findings, the study incorporated process evaluation through heart rate monitoring during flag football sessions. The observed mean heart rate of approximately 143 beats per minute, corresponding to about 71% of predicted HRmax, indicates that the programme was delivered at a broadly moderate to moderately vigorous intensity within the constraints of routine physical education. This level of physiological load is comparable to that reported for other school-based games and small sided sport formats, which typically produce intermittent moderate to vigorous responses rather than sustained near maximal exertion, and supports the feasibility of achieving meaningful training exposure in a class setting [29]. Heart rate values across sex and BMI subgroups are presented descriptively and should be interpreted cautiously. This intensity profile aligns with the movement demands of the intervention, characterized by short high effort bursts, frequent transitions, and reactive play, which plausibly underpin improvements in sprint speed and rope skipping performance. However, because heart rate monitoring was not performed in the control group, it was not possible to determine whether the observed effects reflected the specific movement demands of flag football or, in part, differences in exercise intensity relative to usual physical education. At the same time, the reliance on average session intensity may help explain the absence of a clear group level advantage in shuttle based aerobic fitness, because moderate mean heart rate can coexist with limited accumulated time at vigorous thresholds that more strongly drive cardiorespiratory adaptation.

Key strengths of this trial include its randomized controlled design embedded within routine school physical education, which enhances internal validity while preserving ecological relevance for real world implementation. In addition, the study used standardized, school relevant fitness assessments and incorporated heart rate-based process monitoring, which strengthens confidence that the intervention was delivered with a meaningful physiological stimulus and supports interpretation of outcome patterns. Analyses were conducted according to the intention to treat principle, with multiple imputation for missing post-intervention outcomes, which helps preserve the benefits of randomization and reduces the risk of bias due to loss to follow up.

Several limitations should be considered. First, the trial was conducted in a single Shanghai public school, and the intervention was delivered within a specific PE timetable and teaching context; therefore, external validity to other regions, school types, and PE delivery models in China may be limited. Second, randomization occurred at the individual level within classes, and children in both arms remained in the same school environment; informal exposure to intervention activities, peer spillover, and teacher level influences could not be fully prevented, potentially attenuating between group contrasts. Third, the intervention and control groups differed in weekly session structure, with two 60-min flag football sessions and three 35-min usual PE sessions. Although total weekly instructional time was broadly similar, differences in lesson frequency and format may have influenced training responses. Fourth, the primary outcomes relied on field tests administered in school settings, and VO2max was estimated from the 20 m shuttle run using a prediction equation rather than measured directly; stopwatch timed 50 m sprints may also introduce measurement variability despite standardized procedures. Fifth, heart rate monitoring was conducted only in the intervention group and in a subsample, so physiological load could not be directly compared with usual physical education. This limits inference about whether the observed effects were attributable to the specific characteristics of flag football or to differences in exercise intensity. In addition, maximal HR was estimated rather than directly measured, which limited precise quantification of time spent above vigorous intensity thresholds. Finally, subgroup analyses by sex and weight status were prespecified but based on relatively small stratum sizes and should be interpreted as hypothesis generating; in addition, multiple imputation assumes data were missing at random, an assumption that cannot be verified. In addition, the 1-min jump rope test does not isolate lower-limb muscular endurance. Performance in this task may also be influenced by coordination, rhythm, neuromuscular control, and cardiorespiratory fitness. Therefore, findings for this outcome should be interpreted as task-specific jump rope performance rather than as direct evidence of improved lower-limb muscular endurance.

Conclusion

In this randomized trial in children aged 9 to 11 years, a 10-week flag football programme delivered within PE improved 1 min jump rope performance and 50 m dash time compared with usual PE. In contrast, 20 m shuttle run performance and estimated VO₂max showed no clear group level differences. Prespecified analyses suggested heterogeneity, with cardiorespiratory fitness gains more apparent in boys with normal weight and rope skipping gains more consistent in girls across weight strata. HR monitoring indicated that sessions were delivered at moderate to vigorous intensity. Flag football appears to be a feasible, curriculum-compatible strategy to enhance sprint speed and task-specific jump rope performance in schools, while highlighting the value of differentiated progression to optimize benefits across subgroups.

Supplementary Information

Supplementary Material 1. ^{(33KB, docx)}

Abbreviations

BMI

Body mass index

HR

Heart rate

HRmax

Maximal heart rate

ITT

Intention to treat

MVPA

Moderate to vigorous physical activity

PE

Physical education

SD

Standard deviation

SPSS

Statistical Package for the Social Sciences

VO2max

Maximal oxygen uptake

Authors’ contributions

JYW analyzed the data and drafted the manuscript. YY contributed to data analysis and critically revised the manuscript. YXL, SHG, TX, XZC, DNL, and XHC contributed to data cleaning. JW, TX, and KYY contributed to data collection. JZ and XHC conceived and designed the study, supervised the implementation of the study, interpreted the data, and critically revised the manuscript. XHC also oversaw the revision process as senior corresponding author. All authors read and approved the final manuscript.

Funding

This study was supported by the Science and Technology Commission of Shanghai Municipality (19080503000), the Shanghai Oriental Talent Program (JS2025098), and the Research and Innovation Grant for Graduate Students, Shanghai University of Sport (Project No. YJSCX-2025–025).

Data availability

The 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 protocol was approved by the Ethics Committee of Shanghai University of Sport (Approval No. 2017037). All methods were carried out in accordance with relevant guidelines and regulations and in compliance with the Declaration of Helsinki. Written informed consent was obtained from the parents or legal guardians of all participating children prior to enrolment, after they had been fully informed of the study aims, procedures, potential risks, and anticipated benefits. In addition, participating children were provided with age-appropriate information about the study and gave their assent before participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.