Efficacy of exercise interventions in injury prevention for track and field athletes: a systematic review and meta-analysis of randomized controlled trials

Abstract

Background

Track and field athletes are at high risk of both acute and overuse injuries, with an incidence ranging from 1 to 30 injuries per 1,000 athletic exposures, due to the diverse biomechanical demands of events like overuse injuries in distance runners, traumatic injuries in jumpers, and upper-limb injuries in throwers. Exercise-based interventions have demonstrated injury-preventive effects in other sports; however, their effectiveness in athletics remains unclear.

Objective

This systematic review and meta-analysis of randomized controlled trials (RCTs) aimed to evaluate the effectiveness of exercise-based interventions in preventing sports injuries among track and field athletes by conducting a comprehensive search of PubMed, Web of Science, Scopus, and SPORTDiscus for articles published between 2012 and 2025. The final search was performed on January 1, 2025.

Methods

A total of 2410 records were retrieved and 10 randomized controlled trials met the inclusion criteria. Studies involving track and field athletes (P) exercise-based injury prevention interventions (I) compared with a control group (C) reporting injury incidence risk or rate (O) and employing an RCT design (S) were included. Study quality was assessed using the PEDro scale. Meta-analysis was conducted using RevMan 5.4 employing a random-effects model to calculate pooled rate differences with 95% confidence intervals. Study quality was evaluated using the Physiotherapy EvidenceDatabase scale and the Cochrane Risk of Bias 2 .0 tool to assess methodological rigor and potential bias.

Results

Ten RCTs were included. Interventions included strength training, neuromuscular training, gait retraining, and flexibility or conditioning programs. Only six trials reported statistically significant reductions in injury incidence. Meta-analysis of three eligible studies showed a significant reduction in injury incidence in the intervention group compared to controls (mean difference: -7.63 injuries per 1,000 h; 95% CI: -12.07 to -3.20; p = 0.0007; I² = 88%). Only three studies were eligible for meta-analysis due to consistent outcome reporting in injuries per 1,000 exposure hours.

Conclusion

Exercise-based interventions, particularly strength, neuromuscular, conditioning, flexibility, and gait retraining programs, show moderate-to-strong evidence for reducing injury incidence among track and field athletes.

Review registration

PROSPERO registration number CRD420251083247.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-025-01442-6.

Introduction

Sports-related injuries are a major global public health concern, contributing significantly to disability, medical costs, and reduced quality of life among athletes of all levels [1]. Epidemiological data show injury rates across sports can range from 1.8 to 6.4 per 1,000 h of exposure, depending on sport type and level of play [2]. These injuries disrupt training, hinder performance, and may lead to long-term musculoskeletal complications [1]. Track and field is an Olympic sport comprising four main categories: track events (including sprints, middle- and long-distance running, relays, hurdling, and steeplechase); field events (such as jumping, long jump, triple jump, high jump, and pole vault and throwing disciplines like shot put, javelin, discus, and hammer); road running events; and combined events, notably the decathlon for men and the heptathlon for women [3]. Athletic events also associated with a range of injury types due to the distinct technical demands and physiological requirements of each discipline [4]. For example, distance runners are commonly affected by overuse injuries such as stress fractures and medial tibial stress syndrome [5], while sprinting and jumping events tend to result in a higher incidence of acute injuries like hamstring strains and ankle sprains [6]. Additionally, evidence suggests that high jumpers and pole vaulters are more susceptible to traumatic injuries such as knee ligament tears and shoulder dislocations, whereas athletes in throwing events frequently experience both overuse (e.g., rotator cuff tendinopathy) and acute injuries (e.g., ulnar collateral ligament sprains), particularly affecting the upper body [7].

Track and field events are associated with a high incidence of injuries, ranging from 1 to 30 injuries per 1,000 athletic exposures [7]. A cross-sectional study covering one Olympic cycle(2012–2016)found that 64% of track and field athletes sustained at least one injury during this period [8]. A systematic review further reported that both injury incidence and prevalence exceed 40% among runners [9]. Thigh muscle strains were particularly common in sprinting (34%,n = 41), jumping (15%, n = 15), and middle-distance running (16%, n= 6) events [9]. The longest recovery times were observed in throwing events, with elbow ligament injuries resulting in up to 36 weeks of recovery and elbow muscle injuries requiring up to 39 weeks [9]. Among sprint athletes, foot ligament injuries were associated with the highest rate of performance reduction, with 100% of affected athletes reporting a decline in their performance following injury [9]. Similarly, an epidemiological study spanning 14 international athletics championships(2007–2018)reported 928 injuries among 8,925 male athletes and 597 injuries among 7,614 female athletes [10]. Injury rates per 1,000 registered athletes were highest in combined events, at 235 for men and 212 for women [4]. Another epidemiological review reported injury incidence among track and field ranging between 3.6 and 3.9 injuries per 1,000 h of practice [11]. NCAA surveillance data from 2010 to 2019 estimate between 3.8 and 4.0 injuries per 1,000 exposures in collegiate U.S. athletes [12]. Separate event‑specific data from collegiate jumpers show an incidence of nearly 8.65 injuries per 1,000 athlete‑exposures [13]. Additionally, a 12-month prospective study of bone stress injuries among athletics athletes reported an incidence of 20% per season, underscoring the burden of overuse injury in elite settings [14]. This high prevalence of sports injuries imposes a significant financial burden on individuals, healthcare systems, and employers, and this burden could be minimized through the implementation of effective injury prevention strategies [15].

The broader literature indicates that exercise-based interventions incorporating neuromuscular training, which typically include a combination of strength, agility, balance, core stability, plyometric, and body weight exercises, have demonstrated beneficial effects on injury incidence in various other sports disciplines [16]. However, although randomized controlled trials (RCTs) specifically examining the impact of such interventions on injury prevention in athletics are limited, synthesizing the available evidence is essential to inform practice and guide future research in this area. In one trial evaluating neuromuscular training among youth female track-and-field athletes, the intervention group reported a markedly lower injury incidence rate compared to the control group [17], particularly for medial tibial stress syndrome. Similarly, another study found that participants who followed a structured foot core strengthening program were substantially less likely to sustain running-related injuries (RRIs) over a 12-month period [18]. Gait retraining programs have also been associated with a significant reduction in the incidence of RRIs [19]. For instance, a 12-month follow-up study reported a considerably lower injury rate in the gait retraining group than in the control group [19].

In contrast, a trial investigating a 4-week preconditioning program among novice runners did not observe notable differences in injury incidence between intervention and control groups [20]. Likewise, a cluster-randomized trial of a 39-week athletics injury prevention program did not demonstrate a clear effect in reducing the proportion of athletes with injury complaints that restricted participation [21]. Additionally, a systematic review and meta-analysis involving nine randomized controlled trials concluded that exercise-based interventions did not significantly reduce the overall risk or rate of RRIs. However, subgroup analysis indicated that supervised interventions may offer greater protective benefits [22].

Although exercise-based interventions are widely recognized for their potential to reduce injury risk among track and field athletes, the existing literature presents conflicting evidence. While numerous studies report positive outcomes, such as reductions in injury risk and incidence, some RCTs, systematic reviews, and meta-analyses on runners, have questioned their overall effectiveness. Importantly, no systematic review or meta-analysis to date has specifically examined the efficacy of exercise-based interventions in preventing injuries among track and field athletes. Furthermore, existing RCTs are disproportionately concentrated in endurance running populations, with a marked scarcity of high-quality trials targeting athletes involved in jumping and throwing events, disciplines that exhibit distinct biomechanical demands and injury patterns. As a result, there is currently limited clarity regarding which types of injury prevention exercises are most applicable and effective for this athletic population. A systematic evaluation of the existing literature is therefore warranted to determine whether exercise-based interventions reduce injury risk and incidence among track and field athletes. Such evidence would be valuable for sports medicine practitioners, coaches, and athletes by informing evidence-based decisions on injury prevention strategies. Consequently, this systematic review and meta-analysis aims to comprehensively assess the current evidence on the efficacy of exercise programs in reducing sports injuries within the track and field population. We hypothesize that exercise-based interventions reduce injury incidence by at least 20% per 1,000 athlete-exposures.

Methods

A systematic review and meta-analysis were performed following the guidelines provided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [23] (Supplementary Material 1). This was prospectively registered with the International Register of Systematic Reviews under PROSPERO registration number CRD420251083247.

Search strategy

A comprehensive literature search was conducted to identify relevant studies published between January 1, 2012, and January 1, 2025. This time frame was chosen to capture contemporary research reflecting recent advancements in exercise science, injury prevention protocols, and training practices relevant to modern-day track and field athletes. The following electronic databases were systematically searched: PubMed^® (U.S. National Library of Medicine, USA), Web of Science^® (v.5.4, Clarivate Analytics, USA), SciVerse Scopus^® (Elsevier, The Netherlands), and SPORTDiscus^® (EBSCOhost, USA). Boolean operators and truncation were used to enhance search sensitivity, and Medical Subject Headings (MeSH) were applied when appropriate. The search strategy was independently developed by the review team based on systematic review guidelines without consultation from an academic librarian. Searches were limited to peer-reviewed studies published in English and conducted on human participants (see Supplementary Material 2 for full operational search strings). The final search was conducted on January 1, 2025. The search yielded 247 records from Scopus, 123 from PubMed, 200 from Web of Science, and 272 from SPORTDiscus, resulting in a total of 842 records prior to deduplication.

Two independent reviewers (KW and RJ) executed the database searches and screened studies in multiple phases based on predefined eligibility criteria. In the initial stage, all records retrieved were combined and duplicates were removed using EndNote X9 (Clarivate Analytics, USA) reference management software. The titles, abstracts, and full texts were screened in successive phases according to predefined inclusion and exclusion criteria. Discrepancies in study selection were resolved through discussion between the reviewers. Additionally, reference lists of the included studies were manually screened to identify any relevant articles not captured in the database searches.

Inclusion and exclusion criteria

Randomized controlled parallel group studies were selected based on the population, intervention, comparison, outcome, and study design (PICOS) strategy [24].

• Population (P): Included studies involved track and field athletes competing at any sporting level, categorized according to McKay et al. [25] into elite, sub-elite, and recreational athletes. These included sprinters, hurdlers, jumpers, throwers, and middle- to long-distance runners. Studies involving non-athletic events, mixed sports groups combining athletics with other disciplines, paediatric populations unrelated to sports, or individuals with comorbid medical conditions unrelated to musculoskeletal injuries were excluded.

• Intervention (I): Eligible interventions consisted of exercise-based programs aimed at preventing sports injuries. These included, but were not limited to, strength training, neuromuscular training, plyometrics, balance training, core stability exercises, warm-up routines, and multicomponent injury prevention programs. Interventions could be delivered in supervised or unsupervised settings. Studies combining exercise with non-exercise interventions (e.g., education sessions, supplements, footwear, orthotics) were excluded unless the exercise component could be isolated and analyzed.

• Comparator/Control (C): Studies were included if the control group received no intervention, usual training, or a non-exercise placebo/sham condition.

• Outcomes (O): The primary outcomes included injury incidence, injury risk, or injury rate during the follow-up period. Injuries had to be sports- or training-related and cause either time loss, performance limitation, or require medical attention. Studies reporting only biomechanical parameters, physiological markers, or performance outcomes without injury-related endpoints were excluded. Only studies that reported injury incidence or rate in standardized exposure-based units (e.g., injuries per 1,000 athlete-exposures or per 1,000 h of exposure) were considered eligible for inclusion in the quantitative meta-analysis. Studies reporting injury proportions or relative risk without a common denominator were summarized narratively.

• Study design (S): Only randomized controlled trials (RCTs) with a parallel-group design involving human participants were included, as these are considered the gold standard for evaluating the efficacy of interventions and provide the highest level of evidence for causal inference. Studies had to report on the preventive effect of exercise-based interventions on injury occurrence. Non-randomized designs (e.g., cohort, case-control, cross-sectional studies), studies without sufficient outcome data, case reports, conference abstracts, editorials, reviews, animal, or in vitro studies, and unpublished or grey literature were excluded. Only articles published in English were included due to limited access to translation resources and the necessity for accurate data interpretation during screening and data extraction.

Data extraction

One investigator (KW) independently extracted and tabulated key study characteristics using a standardized form. Extracted variables included: first author, year of publication, country of origin, study design, Physiotherapy Evidence Database (PEDro) score (out of 10), type of sport, athlete level (e.g., recreational, collegiate, elite), gender, mean or median age of participants, total sample size (n), intervention type, frequency of sessions (e.g., sessions/week), duration of the intervention (e.g., number of weeks), total exposure hours (if reported) and outcome metrics (e.g., injury incidence, injury rate, injury risk). In studies reporting injury incidence, we extracted the unit of measurement (e.g., per 1,000 athlete-exposures or per 1,000 person-hours). Athlete-exposure was defined as one athlete participating in one practice or competition [26], while person-hours referred to the cumulative duration athletes were exposed to risk during training or competition [27]. A second investigator (RJ) cross-checked all extracted data for accuracy. Any disagreements between reviewers were resolved through discussion until consensus was reached. Data were extracted in a manner that allowed for comparison of outcomes between the intervention and control groups rather than within-group pre-post comparisons. In cases where interventions included multiple components, only studies in which the effect of the exercise component could be clearly isolated were included. Composite interventions that combined exercise with non-exercise elements were excluded unless the exercise effect could be independently analyzed. The complete data extraction file used for meta-analysis, formatted in RevMan, has been included as Supplementary Material 3. Missing or incomplete outcome data were addressed by contacting study authors via email when key variables were unavailable. If no response was received within four weeks, and data remained unobtainable, those outcomes were excluded from quantitative synthesis and described narratively. Sensitivity analyses were not conducted due to the limited number of studies with missing data.

Assessment of quality

The methodological quality of included randomized controlled trials was evaluated using both the Physiotherapy Evidence Database (PEDro) scale and the Cochrane Risk of Bias 2.0 (RoB 2.0) tool. PEDro scale [28] assesses internal validity and the statistical interpretability of clinical trials and ranges from 0 to 10 points based on criteria such as random allocation, concealed allocation, baseline comparability, blinding, adequate follow-up, and intention-to-treat analysis. Based on the total score, studies were classified as high quality (≥ 7), moderate quality(5–6), or low quality (< 5) [28]. Two investigators (KW and RJ) independently assessed study quality, and discrepancies were resolved by consensus (Supplementary Material 4). In parallel, the RoB 2.0 tool [29] was used to assess domain-specific risk of bias across five key areas:(1) randomization process, (2) deviations from intended interventions, (3) missing outcome data, (4) measurement of the outcome, and (5) selection of the reported result. Each domain was rated as “low risk,” “some concerns,” or “high risk” to provide a more nuanced view of bias affecting intervention effect estimates (Supplementary Material 5).

This dual approach allowed us to assess both the methodological rigor (via PEDro) and the potential influence of bias on study outcomes (via RoB 2.0). Two investigators (KW and RJ) independently conducted the assessments, and discrepancies were resolved by consensus.

Data analysis

Meta-analysis was conducted using Review Manager (RevMan) software, version 5.4 (Cochrane Collaboration, Oxford, UK). The analysis aimed to assess the pooled effect of exercise-based interventions on injury incidence among track and field athletes. The three studies included in the meta-analysis evaluated exercise-based interventions aimed at reducing running-related injuries among track and field athletes. Although the specific intervention components varied ranging from strength and neuromuscular training to conditioning programs, all targeted lower-limb injury prevention and used comparable outcome measures (injuries per 1,000 exposure hours). Thus, pooling was considered methodologically appropriate. Differences in training duration and frequency were accounted for by applying a random-effects model to accommodate expected clinical heterogeneity. The effect size was calculated as the standardized mean difference (SMD), using Hedges’ g with 95% confidence intervals (CIs), which is appropriate for continuous outcomes such as injury incidence rates expressed per 1000 h of exposure. Given the expected heterogeneity in intervention types, durations, and athletic populations, a random-effects model was employed, as this approach accounts for both within-study and between-study variance and is recommended when methodological and clinical diversity are anticipated across studies. The restricted maximum likelihood (REML) method was used to estimate the between-study variance (τ²), which is preferred due to its lower bias in small sample sizes. Statistical heterogeneity was assessed using the Chi-squared (χ²) test and quantified with the I² statistic. An I² value of ≥ 50% was considered indicative of moderate-to-substantial heterogeneity, and a p-value less than 0.05 was considered statistically significant [30]. Subgroup analyses were not performed as all included studies focused on running-related athletic disciplines (e.g., sprinting, middle-distance, or long-distance running), and no jumping or throwing events were represented. Due to the limited number of studies eligible for meta-analysis and the presence of considerable heterogeneity, sensitivity analyses (e.g., omitting the largest-weight study) were not conducted, as they were unlikely to provide meaningful insights into the robustness of the effect estimates. Publication bias was assessed using funnel plot asymmetry and Egger’s regression test when ≥ 10 studies were available for a given outcome, in line with current recommendations to avoid low power in tests of asymmetry. Begg’s test was not used due to its lower sensitivity compared to Egger’s test. The trim-and-fill method was also not applied, as the number of studies per outcome was limited and the presence of high between-study heterogeneity could compromise the reliability of this approach.

Results

The database search resulted the following number of records: SciVerse Scopus^® (247), PubMed^® (123), Web of Science^® (200), and SPORTDiscus^® (272). After removing duplicates, 644 unique articles remained for screening. Title and abstract screening led to 38 articles being selected for full-text review. Following a detailed assessment of the full texts, eight studies satisfied all inclusion criteria. Additionally, two more relevant studies were identified through manual searches, bringing the total number of studies included in the review to 10. Of these, only three trials were included in the meta-analysis. An overview of the search and selection process is illustrated in Fig. 1.

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-analyses flow diagram detailing the review filtering process [23]

This systematic review and meta-analysis included 10 RCTs published between 2012 and 2024 (Table 1). These studies were conducted in various countries: one study each from Iran [30], Chile [17], China [31], Canada [32], the United States of America [33], Netherlands [20], and Hong Kong [19], France [21], Finland [34]and Brazil [18]. All the trials are parallel-group RCTs. Among them, five studies are single-blinded [17–19, 21, 34], while the remaining six are unblinded [20, 30–33]. Five trials [17–20, 34]showed ‘high quality of the selected trials while the other five [4, 30–33] were ‘moderate’ quality according to the PEDro scale (see Supplementary Material 4).

Table 1.

Randomized controlled trials investigating exercise-based interventions for injury prevention in track and field athletes

Author,Year,Country	Study design,PEDro score(out of 10)	Sport,Level	Gender,Age (years)	Sample size (number),Intervention program		Frequency,Time-period	Results
				IG	CG
1. Baltich et al. 2017,Canada [32]	R, UB, P	Running,Recreational	M, F,Median age: IG1: 33, IG2: 30, CG: 31	86 (IG1: 43, IG2: 43),IG1: Functional strength training (lunge, squat, hop, single leg standing, jumps),IG2: Resistance training (Resistance band, isometric against wall)	43,RT	3–5 times/W for 8 W (training), 2 times/W for 16 W (maintenance)	-No significant difference in overall injury incidence: IG1: 21/44 (48%), IG2: 16/44 (37%), CG: 15/41 (35%)-Total running exposure: IG1: 699.4 h → injury rate 32.9 injuries/1000 h, IG2: 538.5 h → injury rate 31.6 injuries/1000 h, CG: 600.3 h → injury rate 26.7 injuries/1000 h-Severity scores of injuries (median): IG1: 43.0, IG2: 37.7, CG: 37.0-Time loss injuries: i. Injuries causing full time loss ≥1 week: IG1: 7, IG2: 4, CG: 1 ii. Injuries requiring moderate/major reduction in running: IG1: 3, IG2: 3, CG: 4
2. Bredeweg et al. 2012,Netherlands [20]	R, UB, P	Running,Recreational	M, F,IG: 39.0 ± 10.7, CG: 37.2 ± 10.9	211,4-week preconditioning (walking + hopping)	221,RT	Walking (50–80 min/W)+Hopping (5 reps/W),4 W	-No significant difference in overall injury incidence: IG: 15.2% (26/171 participants), CG: 16.8% (32/191 participants); p=0.69-Injury rates per 1000 h: IG: 31.0 RRIs/1000 h, CG: 30.0 RRIs/1000 h-Total injuries: IG: 26, CG: 32
3. Chan et al. 2018,Hong Kong [19]	R, SB, P,6	Running,Recreational	M, FIG: 33.6 ± 9.5; CG: 34.2 ± 9.5	166,2-week gait retraining using real-time visual feedback to reduce impact loading	154,RT	4 sessions per W (↑ from 15 to 30 min over the 8 sessions),2 W	-Overall injury incidence at 12 months: IG = 16%, CG = 38%; HR = 0.38; p <0.001-Injury types: i. Achilles tendinitis (IG: 18% vs CG: 0%) ii. Calf strain (IG: 18% vs CG: 0%) iii. Hamstring strain (IG: 11% vs CG: 13%) iv. Iliotibial band syndrome (IG: 11% vs CG: 13%) v. Plantar fasciitis (IG: 7% vs CG: 38%) vi. Patellofemoral pain (IG: 14% vs CG: 29%)vii. Shin splints (IG: 11% vs CG: 2%)viii. Patellar tendinitis (IG: 7% vs CG: 0%) ix. Meniscal injury (IG: 3% vs CG: 5%).
4. Edouard et al. 2021,France[21]	C-R, SB, P,5	Athletics,SE	M, F,IG: 30.1±6.7, CG: 30.2±6.6	449,Core stability, postural control, pelvic/hamstring/lower-leg strengthening and stretching exercises	391,RT	2 sessions per W,39 W	-No significant difference in injury complaints leading to participation restriction: IG: 64.7% vs CG: 65.0%-No significant differences in injury burden: IG: Mean 343.3 days lost/1000 h, CG: Mean 285.6 days lost/1000 h.
5. Leppänen et al. 2024, Finland [34]	R, SB, P,5	Long-distance running,Recreational	M, F,IG1: 39.9±8.7, IG2: 40.6±8.5, CG: 39.9±9.3	219 (IG1: 108; IG2: 111),IG1: Hip and core strengthening exercises, IG2: Ankle and foot strengthening and neuromuscular control exercises	106,RT	Supervised 2×/W + 1–2×/W, home-based (2–4×/W total),24 W	-Time-loss LE injuries: IG1: 60%, IG2: 70%, CG: 80%; IG1vs CG: p=0.044, IG2vs CG: p=0.800-Overuse LE injuries (average weakly prevalence): IG1: 9.2%, IG2: 12.0%, CG: 15.5%; IG1vs CG: p=0.032, IG2vs CG: p=0.320-Substantial overuse LE injuries: IG1: 3.3%, IG2: 5.0%, CG: 7.7%; IG1vs CG: p=0.021, IG2vs CG: p=0.181-Acute LE incidence % (injuries/1000 h): IG1: 10.19% (2.5 injuries/1000 h), IG2: 20.72% (5.3 injuries/1000 h), CG: 3.77% (1.06 injuries/1000 h); IG1vs CG: p=0.224, IG2vs CG: p=0.023-Acute time-loss LE incidence rates (injuries/1000 h): IG1: 2.29, IG2: 4.9, CG: 0.53; IG1vs CG:p=0.071, IG2vs CG: p=0.017
6. Letafatkar et al. 2019,Iran [30]	R, UB, P,6	Running,SE	M,IG: 14.86 ± 0.5, CG:1 4.74 ± 0.3	33 (IG1: 16; IG2: 17),IG1: Conditioning training (strengthening, proprioception, stretches, treadmill runningIG2: Conditional training+visual feedback (mirror) and verbal cues targeting safer running biomechanics	16,RT	3 sessions/W; running task progressed from 15 to 30 min per session,8W	-Overall injury incidence: IG1: 42% (53.50 injuries/1000 hours), IG2: 23% (44.44 injuries/1000 hours), CG: 67% (52.83 injuries/1000 hours); IG1 vs CG: p>0.05, IG2 vs CG: p<0.05-Injury types: i. Patellofemoral pain: IG1: 7 → 6 (14.3% ↓), IG2: 5 → 1 (80% ↓), CG: 6 → 7 (16% ↑) ii. Iliotibial band syndrome: IG1: 7 → 6 (14.3% ↓), IG2: 6 → 3 (50% ↓), CG: 7 → 8 (14% ↑) iii. Shin splints: IG1: 6 → 5 (16.7% ↓), IG2: 6 → 0 (100% ↓), CG: 6 → 7 (16% ↑) iv. Patellar tendinitis: IG1: 4 → 3 (25% ↓), IG2: 4 → 0 (100% ↓), CG: 4 → 4 (no change) v. Low back pain: IG1: 6 → 5 (16.7% ↓), IG2: 6 → 0 (100% ↓), CG: 5 → 5 (no change) vi. Hip pain: IG1: 5 → 4 (20% ↓), IG2: 5 → 0 (100% ↓), CG: 5 → 5 (no change)
7. Mendez-Rebolledo et al. 2021,Chile [17]	R, SB, P,6	Sprinting, middle/long-distance running,E	F,IG: 15.0 ± 2.7, CG: 15.3 ± 2.1	11,Neuromuscular training (jumps, landings, running, strength, agility, balance, core training; based on plyometrics and bodyweight exercises)	11,RT	3 sessions/W,6 W	-Overall injury incidence: IG: 7% (6.58/1000 h), CG: 18% (17.89/1000 h); p <0.05)-Thigh muscle strain incidence: 27.6% ↓ in IG (IG: 2.47/1000 h, CG: 3.41/1000 h)-Knee bursitis incidence: 51.6% ↓ in IG (IG: 1.65/1000 h, CG: 3.41/1000 h)-Knee tendinopathy incidence: 51.8% ↓ in IG (IG: 0.82/1000 h, CG= 1.70/1000 h)-Medial tibial stress syndrome incidence: 86.2% ↓ in IG (IG= 0.82/1000 h, CG= 5.96/1000 h; p <0.05)-Ankle sprain incidence: 75.9% ↓ in IG (IG= 0.82/1000 h, CG= 3.41/1000 h)
8. Taddei et al. 2020,Brazil [18]	R, SB, P,6	Long distance running,Recreational	M, FIG: 40.5± 7.9, CG:41.3 ±6.8	57,Foot and core strengthening protocol focused on intrinsic foot muscles	61,RT	Supervised 1 session/W for 8 W, remote training 3 sessions/W for 12 M	-Overall injury incidence: IG =14% (6.59 injuries/1000 h), CG =32.8% (17.05 injuries/1000 h); p=0.027-Risk of injury: 2.42× higher in CG than IG (HR = 2.42; p=0.035)-Time to injury: Longer in IG (10.15 ± 2.69 months) vs CG (7.63 ± 2.60 months)
9. Toresdahl et al. 2020,USA [33]	R, UB, P,6	Marathon running,Recreational	F,IG: 35.4± 9.1, CG: 36.3±9.8	352,10-min self-directed strength training targeting hip abductors, quadriceps, and core; video- and text-based instructions	368,RT	3 sessions/W,12 W	- No significant difference in overuse injury incidence: (IG: 7.1%, CG: 7.3%, p = 0.90)-Minor injury incidence: No significant difference; 8.3% ↓ in IG (IG: 46.3%, CG: 50.5%, p = 0.26)-Minor injury during race: No significant difference (IG: 14.7%, CG: 16.1%; p= 0.62).
10. Wan et al. 2021,China [31]	R, UB, P,6	Sprinting,Recreational	M,IG: 20.6±1.6, CG: 20.9±1.9	10,Flexibility intervention: static, dynamic, PNF stretching targeting hamstrings	10,Strengthening intervention: concentric and eccentric hamstring strengthening exercises (Nordic hamstring curls, physio-ball curls)	3 sessions/W,8 W	-Hamstring strain risk (peak musculotendinous strain): ↓ in IG (IG: semitendinosus= 0.06 ±0.04 pre vs post p =.004; semimembranosus= 0.05 ±0.05; pre vs post p =.002; biceps long head=0.06±0.05; pre vs post p =.004; all ↓ by 0.03–0.04 FL units), CG: semitendinosus= 0.05 ±0.05 pre vs post p = 0.017; semimembranosus= 0.04 ±0.03; pre vs post p =0.011; biceps long head=0.05±0.04; pre vs post p = 0.009-Hamstring flexibility: in IG (Δ = +4.2°, p =.037), ↑ in CG (Δ = +17.8°, p = 0.001)-Hamstring strength: ↑ in IG (Δ = +9.8 Nm, p = 0.001); No change in CG (p =0.393)

Abd-Add Abductor-Adductor, CG Control Group, C Crossover, C/L Contralateral, CL Control Leg, C-R Cluster-Randomized Trial, F Female, h hours, IR Injury Risk, IG Intervention Group, IG1 Intervention Group 1, IG2 Intervention Group 2, LE Lower Extremity, M Male, M Months, P Parallel, PEDro Physiotherapy Evidence Database, R Randomized Trial, RRI Running related injuries, RT Regular Training, SB Single Blinded, UB Un Blinded, W Weeks

The methodological quality of included studies assessed by the PEDro scale is presented in Supplementary Material 4. According to the Cochrane RoB 2.0 assessment (Fig. 4 and Supplementary Material 5), five studies [11–14, 24]were rated as having a low risk of bias across all domains. Four studies [15, 20, 22, 23]showed “some concerns,” primarily due to deviations from intended interventions and limitations in blinding or outcome measurement. One study [21] was rated as having a high risk of bias due to concerns across multiple domains, including the randomization process and selection of reported results. These findings are detailed in Supplementary Material 5 (Fig. 2).

Fig. 4.

Funnel plot of comparison of effect of exercise-based interventions in reducing injury incidence (injuries per 1000 hours)

Fig. 2.

Summary of the risk of bias across the included randomized controlled trials according to the Cochrane RoB 2.0 tool

The majority of trials focused on recreational-level runners, with a few studies targeting sprinting, middle-distance, or long-distance running, and one involving general track and field athletes. Participant age ranged from 14 years [30]to 41 years [18]. In six studies [18–21, 32, 34]both male and female athletes were represented, while two trials [30, 31]included only male participants and two trials [17, 33] included only female participants.

Sample sizes varied considerably from as few as 20 participants [31]to 840 [21], and intervention duration ranged from two weeks [19]to 39 weeks [21]. Studies utilized a variety of exercise programs as part of their injury prevention interventions. The majority of studies (six out of ten) implemented multi-joint or single-joint strength training [18, 31–34]incorporating resistance band exercises, isometric wall drills, and targeting hip abductors, hamstrings, quadriceps, pelvic muscles, core, and the foot and ankle, particularly the intrinsic foot muscles. One study incorporated core and trunk-focused exercises, specifically targeting postural control and pelvic stability in addition to strength training [21]. Stretching and mobility exercises were used in one trial, comprising static, dynamic, and proprioceptive neuromuscular facilitation (PNF) stretching focused on the hamstrings (n= 10), alongside general flexibility routines [31]. A separate study exclusively evaluated the effectiveness of a neuromuscular training program, which included jumping, landing, running, strength, agility, balance, and core exercises based on plyometrics and bodyweight movements [17]. Another trial assessed the efficacy of a gait retraining intervention, utilizing real-time visual feedback to encourage safer running mechanics and reduce impact loading [19]. Two studies focused on conditioning-based training. Of these, one trial [30]investigated the impact of a comprehensive conditioning package that included strengthening, proprioception, stretching, and treadmill running with or without visual feedback (mirror) and verbal cueing to promote safer running biomechanics while the other study included a preconditioning programme involving walking and hopping [20]. The frequency of sessions also varied, with some interventions involving as few as one session per week [18], while others incorporated up to 5 sessions per week [32].

In terms of injury outcomes, five studies [17–19, 30, 34]reported a statistically significant reduction in injury incidence in the intervention groups compared to controls. For instance, Mendez-Rebolledo et al. [17]demonstrated a 75–86% reduction in ankle and tibial stress injuries through neuromuscular training, while Chan et al. [19]reported a hazard ratio of 0.38 for overall injury incidence favouring the gait retraining group. Building on these findings, Taddei et al. [18]identified a significant reduction in injury risk resulting from foot-core training, with a hazard ratio of 2.42 favouring the intervention group. Meanwhile, Leppänen and colleagues [34]observed a significant reduction in time-loss lower extremity injuries and overuse lower extremity injuries through hip and core strengthening exercises. Letafatkar and colleagues [30]showed the superior effects of a comprehensive conditioning training package with visual feedback and verbal cues in significantly reducing overall injury incidence. In addition, Wan et al. [31]demonstrated significant improvements in hamstring strain risk, flexibility, and strength following concentric and eccentric hamstring strengthening exercises. Conversely, four other trials [20, 21, 32, 33] showed no statistically significant differences in overall injury incidence, although they noted trends in reduced injury severity or time-loss injuries. Across studies, certain intervention types appeared to be associated with more favourable outcomes. Neuromuscular and foot-core training interventions were consistently effective in reducing injury risk, particularly among recreational runners. Similarly, gait retraining interventions showed promising results in altering running mechanics and lowering injury incidence. Strength and conditioning programs targeting lower limb and trunk muscles also demonstrated positive outcomes, although results varied depending on adherence and training intensity. In terms of competition level, most studies focused on recreational or amateur-level athletes, with limited representation of elite or professional track and field participants. This limits the generalizability of findings to higher-level competitors.

The effectiveness of exercise-based interventions in reducing injury incidence in injuries per 1000 h was investigated in three studies [17, 18, 30], and these studies were included in the meta-analysis. A random-effects analysis was conducted due to high heterogeneity observed among the included studies (I² = 88%,p < 0.0001). The pooled analysis demonstrated a statistically significant reduction in injury incidence in the intervention groups compared to the control groups, with a mean difference of −7.63 injuries per 1000 h (95% CI: −12.07 to −3.20, p = 0.0007), favouring the intervention (Fig. 3). These findings provide moderate-to-strong evidence that exercise-based interventions such as strength training, neuromuscular exercises and conditioning training can effectively reduce injury incidence among track and field athletes. Funnel plots revealed a symmetrical distribution of studies around the overall effect size, suggesting no evidence of publication bias (Fig. 4). As fewer than 10 studies were included in the meta-analysis, Egger’s regression test was not conducted, in accordance with recommendations suggesting reduced reliability in small samples.

Fig. 3.

Forest plot of comparison of effect of exercise-based interventions in reducing injury incidence (injuries per 1000 hours)

Discussion

To the best of our knowledge, this is the first systematic review and meta-analysis specifically aimed at evaluating the effectiveness of exercise-based interventions in preventing injuries among track and field athletes. Overall, six out of ten included trials reported that exercise-based strategies such as strength training, neuromuscular exercises, and conditioning programs were more effective than control conditions in reducing injury incidence, injury risk, and improving muscle strength and flexibility. In contrast, the remaining four trials did not report statistically or clinically meaningful reductions in overall injury incidence, although some observed trends toward reductions in injury severity or time-loss injuries. The lack of statistically significant findings in some trials may be attributed to several methodological and implementation-related factors. Firstly, the dosage and duration of interventions in these studies may have been insufficient to elicit meaningful adaptations in musculoskeletal resilience. For example, some trials implemented short-term or low-frequency programs, which may not provide adequate stimulus for neuromuscular or strength gains. Secondly, inconsistent adherence to the intervention protocols, often unreported, may have diluted the intended effects. Third, variation in the baseline fitness or injury risk profile of participants, particularly among novice versus experienced runners, could influence how responsive individuals are to preventive strategies. Finally, the lack of standardized outcome measures and inconsistent definitions of injury across studies may have contributed to mixed results. These factors highlight the need for more robustly designed and adequately powered RCTs with standardized protocols and sufficient intervention exposure.

Our meta-analysis revealed a statistically significant reduction in injury incidence in the intervention groups compared to control groups, with a mean difference of −7.63 injuries per 1000 h of exposure (95% CI: −12.07 to −3.20, p = 0.0007), supporting the potential of exercise-based interventions to effectively reduce injury incidence in this athletic population. This reduction of 7.63 injuries per 1000 h is clinically meaningful, especially in high-volume training environments such as track and field, where athletes may accumulate several hundred training hours over a season. This equates to potentially avoiding multiple injuries per team per season. Interventions that focused on strength training and neuromuscular control appeared particularly effective. These findings support the utility of incorporating structured, preventive exercise programs into regular training to improve athlete health and reduce performance disruptions.

These findings are consistent with those of a previous systematic review and meta-analysis that examined the impact of injury prevention programs (IPPs) on injury incidence rates in team sports [35]. That review reported an average reduction of approximately 35% in overall, lower extremity, thigh, knee, and ankle injury incidence. It also found that most IPPs produced statistically significant reductions in the risk of overall and lower limb injuries when compared to control conditions. Notably, interventions focused solely on strength training (IRR = 0.30, 95% CI: 0.10-0.93) or flexibility (IRR = 0.49, 95% CI: 0.36–0.68), as well as those incorporating stability exercises, were identified as the most effective approaches for reducing injury rates in youth team sports [35]. Another systematic review and meta-analysis also evaluated the impact of exercise-based prevention programs on the risk of RRIs among endurance runners [22]. The pooled findings indicated that supervised interventions were associated with a significantly lower injury risk in the intervention groups compared to control groups (z =−3.75, p< 0.001). FurtherSalam et al. pooled data from 24 studies and concluded that sports injury prevention interventions were effective in significantly reducing injury rates [36]. Specifically, they observed reductions in overall injury incidence (RR = 0.66, 95% CI: 0.53–0.82), injuries per hour of exposure (RR = 0.63, 95% CI: 0.47–0.86), and injuries per number of exposures (RR = 0.79, 95% CI: 0.70–0.88). Two meta-analyses conducted by Lauersen et al. [37, 38]concluded that strength training is effective in reducing the risk of sports-related injuries, which aligns with the findings of the current review, where strength-based interventions also demonstrated promising effects. Furthermore, one of these analyses found that proprioceptive training (RR = 0.55, 95% CI: 0.35–0.87) and multicomponent exercise programs (RR = 0.66, 95% CI: 0.52–0.83) were also significantly effective in lowering injury risk [37].

Another review focused on adult recreational athletes (Smith et al., 2023) reported modest risk reduction (RR = 0.94) with high heterogeneity (I² = 81%) and limited statistical strength [39]. Meanwhile, an integrated neuromuscular training analysis observed substantial reductions in lower-limb injuries (SMD = 0.68) and performance gains in jumping tasks (SMD = 0.45) [40]. In contrast, our review focused specifically on track and field athletes and found stronger effect, particularly with foot‑core and neuromuscular training, likely due to standardized sport-specific protocols, clearer injury definitions, and constrained intervention heterogeneity.

The observed reduction in injury incidence can be attributed to several physiological and biomechanical mechanisms by which exercise-based interventions exert their protective effects. Strength training, for instance, enhances the force absorption capacity of muscles and tendons, reducing joint loading and improving neuromuscular control, which is critical in high-impact sports like track and field [41]. Improved muscular strength also contributes to joint stability and optimized movement mechanics, thereby reducing the likelihood of non-contact injuries [41]. This was demonstrated by a scoping review of randomized controlled trials, which found that strength training effectively reduces sports injury incidence while simultaneously enhancing performance measures such as sprint speed, jump height, and muscular endurance [42]. In addition, neuromuscular training programs, which often include balance, plyometric, and proprioceptive exercises, are thought to reduce injury risk by enhancing sensorimotor function and postural control [43]. This is particularly relevant for preventing lower extremity injuries by improving coordination and reactive stability during high-speed or multidirectional activities common in sprinting, hurdling, and jumping events [43]. This is further supported by a systematic review that demonstrated integrated neuromuscular training significantly improved jump performance (SMD = 0.45, 95% CI: 0.30–0.60,p < 0.001) and reduced the incidence of lower limb injuries (SMD = 0.68, 95% CI: 0.62–0.74, p< 0.001) [40]. Flexibility and mobility training may contribute by improving joint range of motion and muscle-tendon unit compliance, which help in dissipating mechanical stress during dynamic athletic movements [44]. Enhanced flexibility may also reduce passive muscle stiffness and improve stride efficiency, particularly in running-based disciplines [44]. A systematic review and meta-analysis of randomized controlled trials, also reported that that active stretching significantly improved maximal isometric strength and ankle dorsiflexion range of motion while reducing lower limb and trunk injuries in athletes [45]. Additionally, conditioning training that include core stability and dynamic stabilization exercises contribute to a more efficient kinetic chain, reducing compensatory movements that may predispose athletes to overuse injuries [46]. Therefore, improving athletes’ muscular strength, power, and flexibility may contribute to reducing the incidence of various sports injuries.

While the findings suggest that strength training and neuromuscular control exercises may be particularly effective in reducing injury incidence, the limited number of included studies especially those eligible for quantitative synthesis, precludes definitive conclusions about the superiority of specific intervention types. The small sample size and variability in intervention protocols across studies may introduce bias and reduce the precision of subgroup-level inferences. Therefore, these results should be interpreted with caution, and future research should aim to conduct adequately powered head-to-head comparisons of different exercise-based modalities to determine the most effective strategies for injury prevention in track and field athletes.

Strengths and limitations

This systematic review and meta-analysis demonstrate several key strengths. First, the study adhered to rigorous methodological standards by following the PRISMA guidelines and prospectively registering the protocol with PROSPERO, ensuring transparency and reducing bias. A comprehensive and systematic literature search was conducted across multiple major databases including PubMed, Web of Science, SciVerse Scopus, and SPORTDiscus using carefully constructed search strategies that incorporated MeSH terms and Boolean operators to maximize retrieval sensitivity. Manuscripts were further identified through reference list checks. The inclusion of only RCTs with parallel-group designs enhanced the internal validity of the findings. Furthermore, data extraction and quality assessment were independently performed by two reviewers, with discrepancies resolved by consensus, thus strengthening the reliability of the data synthesis. The use of the PEDro scale and Cochrane RoB 2.0 tool provided a standardized assessment of study quality, and the application of random-effects meta-analysis accounted for heterogeneity across studies. Finally, the manual screening of reference lists supplemented database searches to minimize the risk of missing relevant studies. Importantly, this review is the first to focus exclusively on track and field athletes, thereby providing sport-specific insights and addressing a gap in the literature where prior reviews often combined multiple sports or athletic populations. This specificity allows for more precise and applicable recommendations tailored to the unique injury profiles and training demands of track and field disciplines.

Despite its methodological rigor, this review is subject to certain limitations that impact the generalizability and interpretation of its findings. The total number of included studies was relatively small (n = 10), limiting statistical power and the robustness of subgroup analyses. Considerable heterogeneity existed in the interventions employed, participant characteristics (e.g., age, sex, athletic level), and outcome measures, which complicated direct comparisons and may have influenced the pooled effect estimates. Furthermore, high statistical heterogeneity was observed in the meta-analysis (I² = 88%), which limits the certainty of pooled estimates. In addition to statistical heterogeneity, the pooled analysis revealed wide confidence intervals in several studies, reflecting considerable imprecision in the estimated effects. This imprecision may stem from small sample sizes, variability in intervention content, duration, and underlying clinical heterogeneity. Further contributing to the observed heterogeneity were variations in the competitive level of athletes (e.g., youth, collegiate, or elite), the duration and intensity of exercise interventions, and differences in injury definitions and follow-up periods across studies. These factors likely influenced both injury risk and responsiveness to the intervention, thus affecting the consistency of outcome measures. Additionally, only three studies were eligible for quantitative synthesis in the meta-analysis, further restricting the breadth of pooled analysis. Notably, several included trials were unblinded or single-blinded, which may introduce bias and affect internal validity. Moreover, the presence of small-study effects cannot be ruled out, as smaller trials may overestimate intervention effects due to publication bias or methodological shortcomings. Variability in study protocols, including differences in intervention delivery, adherence monitoring, and outcome assessment, adds to the complexity of interpreting results and may contribute to inconsistency across findings. The variation in risk of bias domains, such as randomization methods and selective outcome reporting, further complicates the synthesis of evidence and highlights the need for cautious interpretation of pooled effect sizes. The predominance of studies conducted in recreational athletes, and the scarcity of data from elite competitive settings, limit extrapolation to high-performance athletes. Potential publication bias cannot be completely excluded despite symmetrical funnel plots, as the review was restricted to English-language publications. Lastly, variation in blinding methods and quality scores across studies, with most rated as moderate quality, suggests caution in interpreting the overall strength of the evidence.

Practical recommendations

Based on the findings and identified gaps, several practical recommendations can be proposed for both practitioners and future research. Sports practitioners working with track and field athletes should consider integrating tailored exercise-based injury prevention programs that emphasize strength training, neuromuscular exercises, conditioning and gait retraining with visual feedback as these interventions demonstrated efficacy in reducing injury incidence. Given the heterogeneity of effective protocols, programs should be individualized to target specific muscle groups and movement patterns relevant to the athlete’s discipline and injury risk profile. Where reported, effective programs generally prescribed exercise sessions at a minimum frequency of two to three times per week, with durations ranging from 15 to 45 min per session, sustained over at least 8 to 12 weeks. Practitioners are therefore encouraged to adhere to these minimum thresholds when designing prevention protocols. Future research should aim to conduct larger, high-quality RCTs with standardized intervention protocols and outcome measures, particularly in elite athlete populations, to better determine optimal exercise modalities, frequencies, and durations. Furthermore, future studies should prioritize conducting long-term randomized controlled trials exceeding 12 months to better capture the sustained effects of exercise-based interventions on injury prevention. Special emphasis is needed on elite athlete cohorts, as these populations are underrepresented in current research. Standardizing intervention protocols and outcome measures across studies will be critical to improving comparability and generating robust, generalizable evidence to guide practice at the highest levels of competition. Additionally, combining exercise interventions with biomechanical feedback or gait retraining shows promise and warrants further investigation. Coaches and sports medicine practitioners working with track and field athletes should prioritize implementing tailored exercise-based injury prevention programs focusing on strength training, neuromuscular exercises, conditioning, and gait retraining with visual feedback. Interventions should be delivered at least two to three times per week, with session durations between 15 and 45 min, sustained for a minimum of 8 to 12 weeks. Individualization based on the athlete’s event, injury risk, and performance level is essential. Regular monitoring of athlete adherence and response can optimize program effectiveness and minimize injury risk. It is also recommended that researchers report detailed participant demographics and adherence data to enhance external validity. Finally, ongoing collaboration between sports scientists, physiotherapists, and coaches is critical to ensure that evidence-based injury prevention strategies are effectively implemented within training environments.

Conclusions

The findings of this review indicate that exercise-based interventions specifically strength training, neuromuscular exercises, conditioning training, stretching, and gait retraining with visual feedback are associated with a statistically significant reduction in injury incidence among track and field athletes. The meta-analysis showed an average decrease of 7.63 injuries per 1000 h of exposure (95% CI: −12.07 to −3.20, p = 0.0007) for strength training, neuromuscular exercises, and conditioning interventions. No adverse events related to these interventions were reported, suggesting they are safe and feasible. Given the limited number of high-quality trials and heterogeneity in study designs, further research, particularly in elite athletes is needed to confirm these findings and refine intervention protocols.

Abbreviations

CI

Confidence Interval

MeSH

Medical Subject Headings

NCAA

National Collegiate Athletic Association

PEDro

Physiotherapy Evidence Database

PICOS

Population Intervention Comparator Outcomes and Study

PNF

Proprioceptive Neuromuscular Facilitation

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

International Prospective Register of Systematic Reviews

RCT

Randomized Controlled Trial

REML

Restricted Maximum Likelihood

RoB 2.0

Cochrane Risk of Bias 2.0

RRIs

Running-Related Injuries

SMD

Standardized Mean Difference

USA

United States of America

Authors’ contributions

KW conceived and designed the study. KW and RJ searched databases. KW and RJ were involved in retrieving data. KW and RJ drafted the manuscript. KW, RJ and APH revised the paper. All authors provided critical feedback on the manuscript. All authors read and approved the final manuscript.

Funding

Not applicable.

Data availability

Not applicable.

Declarations

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Conflict of interest

The authors declare no competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.