Abstract
Musculoskeletal injuries have a substantial impact on athletes, affecting sports performance and increasing the risk of future musculoskeletal disorders (e.g., osteoarthritis). Injury prevention programs are essential to reduce the risk of sport-related injuries and meta-analyses can provide a large amount of information in a single article.Objective: To summarize the pooled effects of injury prevention programs focused on any form of physical exercise in the incidence and risk of musculoskeletal injuries and reinjuries by sports and musculoskeletal body regions.Methods: The CINAHL (via EBSCOhost), Embase (via Elsevier), Epistemonikos, PubMed, Scopus, SPORTDiscus (via EBSCOhost), and the Cochrane Library e-databases were searched from inception to 7 October 2024. Systematic reviews with meta-analyses of randomized clinical trials were considered. The methodological quality of systematic reviews was assessed with AMSTAR 2. The degree of overlap between meta-analyses of interest
sports and musculoskeletal body regions.Methods: The CINAHL (via EBSCOhost), Embase (via Elsevier), Epistemonikos, PubMed, Scopus, SPORTDiscus (via EBSCOhost), and the Cochrane Library e-databases were searched from inception to 7 October 2024. Systematic reviews with meta-analyses of randomized clinical trials were considered. The methodological quality of systematic reviews was assessed with AMSTAR 2. The degree of overlap between meta-analyses of interest was calculated.Results: Fourteen systematic reviews were included. Thirteen of these reviews were focused on soccer. Overall, meta- analyses including a specific injury prevention program (FIFA 11+ and FIFA 11+ kids) found that these programs may reduce the risk of musculoskeletal injuries among soccer players. Concretely, FIFA 11+ may reduce the risk of ankle, knee, hip/groin, and hamstring injuries, whereas FIFA 11+ kids may decrease the risk of ankle and knee injuries.Conclusions: FIFA 11+ and FIFA 11+ kids may reduce the risk of sports musculoskeletal injuries, mainly in the lower limbs. However, many clinical and methodological issues (e.g., the lack of meta-analyses in many types of sports) were discussed and highlighted the difficulty of making robust clinical recommendations with the current data. Keywords:athlete; incidence; meta-analysis; musculoskeletal; overview; review; sport 1. Introduction Sport-related musculoskeletal injuries are mainly characterized by contact or non- contact injuries that occur in either practices or competitions and affect muscles, bones, Healthcare2025,13, 1530 https://doi.org/10.3390/healthcare13131530
Healthcare2025,13, 1530 2 of 29 tendons, joints, ligaments, and other soft tissues [1]. When musculoskeletal injuries appear they have a substantial impact on athletes, altering sports performance [2] and fostering future musculoskeletal disorders (e.g., osteoarthritis), as well as mental health issues such as distress, anxiety, and sleep disturbances [3,4]. Musculoskeletal injuries have also important implications in sports competitions, delaying the return to play [5], increasing practice and game time loss [6], and raising the economic burden on athletes, teams, and sporting competitions [7,8]. Sport-related musculoskeletal injuries are highly incident in many sports disciplines and the epidemiology, causes, and factors related to the appearance of injuries can vary greatly between sports, levels of play, sex, and age groups. For example, anterior cruciate ligament injuries may be more incident in female basketball players, whereas male bas- ketball players may have a higher incidence of ankle sprain [9]. Hamstring injuries may represent around 10% of all injuries in field-based team sports such as soccer, rugby union, field hockey, Gaelic football, hurling, and Australian football [10]. Furthermore, shoulder strains and tendinopathies may be very common among softball players [11], and shoulder injuries may have a high probability of becoming chronic in sports such as water polo [12]. In this context, injury prevention programs (IPPs) may be essential to reduce the inci- dence of sport-related musculoskeletal injuries and the risk of sport-related musculoskeletal reinjuries. In sports, IPPs are considered cost-effective, with total cost savings that may reach up to€462 per athlete [13]. Furthermore, IPPs focused on physical exercise have received great interest, and physical exercise is one of the first-line treatments for muscu- loskeletal disorders [14,15]. Programs such as FIFA 11, FIFA 11 +, FIFA 11 + kids, balance board training, the prevention injury and enhance performance program, the bounding exercise program, and neuromuscular training (Knäkontroll) program are mainly based on different types of physical exercises, and some systematic reviews have underlined their positive effects on hamstring injuries, knee injuries, or ankle injuries in sports disciplines such as soccer [16–18]. Recent advances have been made to increase the implementation of IPPs in sports, and an
enhance performance program, the bounding exercise program, and neuromuscular training (Knäkontroll) program are mainly based on different types of physical exercises, and some systematic reviews have underlined their positive effects on hamstring injuries, knee injuries, or ankle injuries in sports disciplines such as soccer [16–18]. Recent advances have been made to increase the implementation of IPPs in sports, and an overview of reviews including more than 100 syntheses of the literature has highlighted the importance of establishing IPPs for helping athletes of different sports [19,20]. The proliferation in the number of systematic reviews in this field is a fact, but many of these systematic reviews have not developed meta-analyses [21,22], or have not performed specific meta-analyses by sports, as well as by body regions [23,24]. To our knowledge, there is no overview of systematic reviews with meta-analyses that has focused on including only those meta-analyses showing results that may have direct clinical applicability. Therefore, a new overview of systematic reviews with meta-analyses on this topic is timely and may help sports clinicians to consume a large amount of critically appraised information in a single article [25,26]. The objective of this overview of systematic reviews with meta-analyses was to sum- marize the pooled effects of IPPs focused on any form of physical exercise (e.g., aerobic training, regular physical activity, or yoga) in the incidence and risk of musculoskeletal injuries and reinjuries by sports and musculoskeletal body regions (e.g., IPPs on ankle injuries in soccer). 2. Materials and Methods This overview of systematic reviews with meta-analyses followed the PRIOR state- ment [27] and the PRISMA statement for abstracts [28]. The review protocol was prospec- tively registered at Open Science Framework:.
Healthcare2025,13, 1530 3 of 29 2.1. Deviations from the Protocol Deviations in the protocol were reported in Supplementary File S1. 2.2. Data Sources and Search Strategy One co-author (JMC) screened the following e-databases: CINAHL (via EBSCOhost), Embase (via Elsevier), Epistemonikos, PubMed, Scopus, SPORTDiscus (via EBSCOhost), and the Cochrane Library from inception to 7 October 2024. Search filters by type of document (conference abstracts were not considered) were imposed when possible. The full search strategy was reported in Supplementary File S2. A manual search was developed to supplement search strategies. We manually screened syntheses of literature related to our scope that were retrieved during search strategies (e.g., scoping reviews or overviews of reviews). 2.3. Eligibility Criteria The PICOS (Population, Intervention, Control, Outcome, Study Design) framework was used to develop the eligibility criteria [29]. Inclusion criteria: P: Athletes without clinical (e.g., duration of musculoskeletal injury), personal (e.g., age groups), and sports (e.g., level of play) restrictions. I: IPPs focused exclusively on physical exercise, regular physical activity, and/or mind–body exercise interventions (e.g., Nordic hamstring exercises). We also included those IPPs where any of these interventions were a core part, although other interventions were considered. C: No restrictions were imposed. O: Pooled incidence or risk of developing sport-related musculoskeletal injuries or reinjuries. We considered both non-contact and contact injuries. Musculoskeletal injuries are defined as those injuries that occur in either practices or matches and affect muscles, bones, tendons, joints, ligaments, and other soft tissues [1]. We focused these outcomes on meta-analyses conducted by sports and musculoskeletal body regions (e.g., IPPs on the incidence of ankle injuries in basketball). If possible, we also considered any subgroup meta-analyses developed by sports and musculoskeletal body regions, but adding other important factors (e.g., sex-differences IPPs ankle injuries in basketball). In terms of age groups, the results were divided into children, adolescents, and adults, when possible. We followed the suggestions of the World Health Organization to consider children and adolescents those participants 19 years old or younger (https://www.who.int/health- topics/adolescent-health/#tab=tab_1, accessed on 10 September 2024). S: Systematic reviews with meta-analyses of randomized clinical trials (e.g., cluster randomized controlled trials) published
basketball). In terms of age groups, the results were divided into children, adolescents, and adults, when possible. We followed the suggestions of the World Health Organization to consider children and adolescents those participants 19 years old or younger (https://www.who.int/health- topics/adolescent-health/#tab=tab_1, accessed on 10 September 2024). S: Systematic reviews with meta-analyses of randomized clinical trials (e.g., cluster randomized controlled trials) published in peer-reviewed journals. We included all studies that were justified as systematic reviews. Meta-analyses evaluating two or more original studies were only considered. Exclusion criteria: [I] Systematic reviews did not specify in their methods that meta-analyses or their subgroups were performed by sports and musculoskeletal body regions and included only randomized clinical trials. [II] Meta-analyses combined different sports injuries rather than musculoskeletal injuries (e.g., concussion). [III] Meta-analyses include studies evaluating army recruits. Although army recruits are sometimes considered to be athletes, we have excluded them since the training environ- ment and objectives of this population are completely different from other athletes (e.g., basketball players).
Healthcare2025,13, 1530 4 of 29 [IV] Conference abstracts and proceedings. [V] No full-text access. This criterion was applied if, after requesting the full text from the authors, we did not receive a response, or they did not send us the full text. [VI] In network meta-analyses, no direct comparisons were reported. 2.4. Study Selection Study selection was independently conducted by two co-authors (JMC and SPE). One co-author (JMC) used Zotero 6.0.36 Citation Management Software to include the references retrieved by e-databases. All references were manually checked, and duplicates were removed. Then, titles and abstracts were read and irrelevant studies regarding the objectives of this overview were excluded. Subsequently, JMC and SPE independently analyzed full texts if abstracts seemed eligible or if abstracts were unavailable. Disagreements between these co-authors were solved by consensus. We consulted with a third co-author (CGM) for the inclusion of two studies [30,31]. The percentage of agreement between JMC and SPE was calculated using the number of studies rated with the same score before pooling the results of their independent assessments. The percentage of agreement was 98.7%. 2.5. Methodological Quality Assessment of Systematic Reviews Two co-authors (SPE and JMS) independently used AMSTAR 2 to analyze the method- ological quality of systematic reviews [32]. This tool is composed of sixteen items that can be evaluated as yes, partially yes, or no. Using the overall score is not recommended [32], but the following items are considered critical: items: 2, 4, 7, 9, 11, 13, 15 [32]. We solved disagreements between SPE and JMS by consensus and we calculated the percentage of agreement between them using the number of items rated with the same score before pooling the results of their independent assessments. 2.6. The Degree of Overlap Between Reviews One co-author (JMC) built matrices of evidence to calculate the corrected covered area (CCA) that is needed to know the degree of overlap between systematic reviews [33]. The CCA is defined as the area that is covered once original studies are removed the first time they are counted. The degree of overlap can be classified as slight (CCA 0–5%), moderate
One co-author (JMC) built matrices of evidence to calculate the corrected covered area (CCA) that is needed to know the degree of overlap between systematic reviews [33]. The CCA is defined as the area that is covered once original studies are removed the first time they are counted. The degree of overlap can be classified as slight (CCA 0–5%), moderate (CCA 6–10%), high (CCA 11–15%), or very high (CCA > 15%) [33]. We only calculated the degree of overlap when at least two systematic reviews meta- analyzed the same sport and musculoskeletal body region (e.g., ankle injuries in soccer). As we only included in this overview those meta-analyses that satisfied our inclusion criteria, we only considered the references included in the meta-analyses of interest to calculate the degree of overlap. Finally, one co-author (CGM) built a bar plot to depict the degree of overlap between systematic reviews. 2.7. Data Extraction Two co-authors (JMC and SPE) extracted independently from each review the follow- ing information when possible: (1) total sample size, (2) total hours of exposure, (3) the number and type of injury, (4) mechanism of injury (non-contact or contact), (5) type of sport, (6) level of play, (7) age groups, (8) sex, (9) location of original studies, (10) defini- tion of injuries, (11) study design of original research, (12) type of IPPs and main type of physical exercise reviewed, (13) observation period of intervention, (14) the percentage of compliance, (15) type of control group. This information was calculated if it was not directly reported and the authors of systematic reviews included sufficient information to calculate it. In addition, we extracted from meta-analyses of interest the following information: (1) statistical metric (e.g., risk ratio), (2)p-value, (3) I-square value, (3) the 95% confidence interval, (4) the number of studies meta-analyzed, (5) total sample meta-analyzed, (6) the
Healthcare2025,13, 1530 5 of 29 certainty of evidence of meta-analysis using the GRADE system. Corresponding authors were not contacted to clarify or report additional information. Disagreements between JMC and SPE were solved by consensus. The percentage of agreement between these co-authors was calculated using the number of items rated with the same score before pooling the results of their independent assessments. The percentage of agreement was 90%. 2.8. Data Synthesis The results are reported in the main text by sports and type of IPP (e.g., FIFA 11). Inside this category, we included all meta-analyses evaluating different musculoskeletal body regions. We also include in this category all subgroups that could be extracted from the reviews included if they satisfy our inclusion criteria. On the other hand, those meta- analyses of interest combining original research that studied different IPPs (e.g., FIFA 11 and a Nordic exercise program) are reported in tables, but they are not shown in the main text. This decision was made because we aimed to show those meta-analyses with the highest clinical replicability, which is essential to translating clinical research into clinical practice. 3. Results A total of 7457 references were retrieved from e-databases. After removing duplicates, 2193 references were read and evaluated by title and abstract. Of them, 315 references were analyzed in full text. Finally, 14 systematic reviews with meta-analyses were included (Figure ). The list of excluded studies with the reasons for exclusion during the analysis in the full text was reported in Supplementary File S3. Additionally, 25 references were manu- ally found, and all were excluded (Supplementary File S4). Table of the included reviews [16–18,34–44]. Figure 1.The PRISMA 2020 flow diagram.
Healthcare2025,13, 1530 6 of 29 Table 1.Characteristics of the included systematic reviews. Study and Year of Publication Population (General Review) Study Design Original Research (General Review) Intervention Group (General Review) Control Group (General Review) Meta-Analyses of Interest and the Certainty of Evidence (GRADE) Al Attar et al., 2022 [17] N: 7828 Hours of exposure: 863.7 The number of injuries: 451 knee injuries Mechanism of injury: non-contact injuries Sport: soccer Level of play: collegiate or amateur Age groups: adolescents and adults (ages ranged from 12 to 45 years) Sex: females and males Location: Africa, Europe, Oceania, and the Americas Did the authors provide a specific definition of injuries? Yes Nine cluster randomized controlled trials Types of IPP: FIFA 11, FIFA 11+, neuromuscular warm-up, pre-training and post-training FIFA 11+ Observation period: interventions ranged from 6 months to 12 months Compliance: It ranged from 47% to 85% Usual care (undefined), Pre-training FIFA 11+ program Knee injuries Overall reduction in risk of knee injuries per 1000 h of exposure in favor of the intervention group (IRR 0.446; 95%CI 0.321–0.619;p= 0.000; I 2 29%; k = 9; N = 813,952) Anterior cruciate ligament injuries Overall reduction in risk of anterior cruciate ligament injuries per 1000 h of exposure in favor of the intervention group (IRR 0.401; 95%CI 0.215–0.750;p= 0.004; I 2 0%; k = 4; N = 557,302) Subgroup analyses by sex: knee injuries Reduction in risk of knee injuries per 1000 h of exposure in males (IRR 0.537; 95%CI 0.355–0.813;p= 0.003; I 2 16.65%; k = 5; N = 354,531) and females (IRR 0.354; 95% CI 0.221–0.565;p= 0.001; I 2 19.88%; k = 3; N = 411,773) The certainty of evidence:the GRADE system was not applied
Healthcare2025,13, 1530 7 of 29 Table 1.Cont. Study and Year of Publication Population (General Review) Study Design Original Research (General Review) Intervention Group (General Review) Control Group (General Review) Meta-Analyses of Interest and the Certainty of Evidence (GRADE) Al Attar et al., 2022b [16] N: 9633 Hours of exposure: 775,606 The number of injuries: 529 ankle injuries Mechanism of injury: UR Sport: soccer Level of play: middle and high school, collegiate, amateur, or elite (e.g., Norwegian First, Second, and Third Division) Age groups: children, adolescents, and adults (ages ranged from 7 to 35 years) Sex: females and males Location: Africa, Europe, Oceania, and the Americas Did the authors provide a specific definition of injuries? Yes Eight cluster randomized controlled trials and one individual randomized controlled trial Types of IPP: FIFA 11, FIFA 11+, FIFA 11+ kids, pre-training and post-training FIFA 11+, soccer-specific neuromuscular training program, targeted exercise program including balance exercise, balance training program Observation period: interventions ranged from 2.5 months to 12 months Compliance: It ranged from 28% to 95/100% Pre-training FIFA 11+ program, standardized warm-up, home-based stretching program, neuromuscular training, Nordic hamstring lowers and groin strength training, resisted running using elastic bands, standard conditioning exercises, without any balance training exercises, aerobic warm-up, static and/or dynamic stretches, and soccer skills practice, small-sided games Ankle injuries Overall reduction in risk of ankle injuries per 1000 h of exposure in favor of the intervention group (IRR 0.64, 95%CI 0.54–0.77;p= 0.000; I 2 0%; k = 9; N = UR) Subgroup analyses by sex: ankle injuries Reduction in risk of ankle injuries per 1000 h of exposure in males (IRR 0.58, 95%CI 0.45–0.76;p= 0.000; I 2 0%; k = 4; N = UR) or trials including males and females (IRR 0.59, 95%CI 0.42–0.83;p= 0.002; I 2 0%; k = 3; N = UR) No differences between groups were observed in females (IRR 0.85, 95%CI 0.59–1.22;p= 0.377; I 2 0%; k = 2; N = UR) Subgroup analyses by type of intervention: ankle injuries FIFA 11 +: reduction in risk of ankle injuries per 1000 h of exposure (IRR 0.64, 95%CI 0.48–0.84;p= 0.002;
0.42–0.83;p= 0.002; I 2 0%; k = 3; N = UR) No differences between groups were observed in females (IRR 0.85, 95%CI 0.59–1.22;p= 0.377; I 2 0%; k = 2; N = UR) Subgroup analyses by type of intervention: ankle injuries FIFA 11 +: reduction in risk of ankle injuries per 1000 h of exposure (IRR 0.64, 95%CI 0.48–0.84;p= 0.002; I 2 36%; k = 5; N = UR) Balance training exercises alone: reduction in risk of ankle injuries per 1000 h of exposure (IRR 0.59, 95%CI 0.41–0.84;p= 0.004; I 2 0%; k = 4; N = UR) The certainty of evidence:the GRADE system was not applied
Description
This review summarizes the effects of injury prevention programs on musculoskeletal injuries in athletes.