Sports Injury Rehabilitation: A Narrative Review of Emerging Technologies and Biopsychosocial Approaches

Abstract

The purpose of this narrative review is to critically appraise recent advances in sports injury rehabilitation—primarily focusing on biopsychosocial (BPS) approaches alongside emerging technological innovations—and identify current gaps and future directions. A literature search was conducted in PubMed, Scopus, and Web of Science for the years 2018–2024. Eligible records were English-language, human studies comprising systematic reviews, clinical trials, and translational investigations on wearable sensors, artificial intelligence (AI), virtual reality (VR), regenerative therapies (platelet-rich plasma [PRP], bone marrow aspirate concentrate [BMAC], stem cells, and prolotherapy), and BPS rehabilitation models; single-patient case reports, editorials, and non-scholarly sources were excluded. The synthesis yielded four themes: (1) BPS implementation remains underutilised owing to a lack of validated tools, variable provider readiness, and system-level barriers; (2) wearables and AI can enhance real-time monitoring and risk stratification but are limited by data heterogeneity, non-standardised pipelines, and sparse external validation; (3) VR/gamification improves engagement and task-specific practice, but evidence is dominated by pilot or laboratory studies with scarce longitudinal follow-up data; and (4) regenerative interventions show mechanistic promise, but conclusions are constrained by methodological variability and regulatory hurdles. Conclusions: BPS perspectives and emerging technologies have genuine potential to improve outcomes, but translation to practice hinges on (1) pragmatic or hybrid effectiveness–implementation trials, (2) standardisation of data and intervention protocols (including core outcome sets and effect-size reporting), and (3) integration of psychological and social assessment into routine pathways supported by provider training and interoperable digital capture.

1. Introduction: Rethinking Sports Injury Rehabilitation

Sports injuries affect not only elite and amateur athletes but also public health and healthcare delivery. The high occurrence of such injuries, their financial burden, and their long-term effects on athletic performance and quality of life have increased scientific interest in recent decades. For example, in the United States, the cost of hospital-treated nonfatal injuries, including sports-related injuries, was estimated at USD 1.853 trillion in 2013, with significant portions attributed to medical spending, work losses, and decreased quality of life [1]. Running-related injuries, for example, can lead to limitations in daily activities and work absenteeism, further impacting public health and economic productivity [2]. Despite the economic toll of sports-related injuries, these expenses need to be placed in a broader context that focuses on the health benefits of exercise. Empirical evidence shows that an increased amount of physical exercise is associated with substantial health advantages, which will also lead to savings related to healthcare expenditures, including a reduction in healthcare spending because of overall health profile enhancement [3]. Data have repeatedly revealed a disparity in rehabilitation care for sports injuries between elite and amateur athletes, as highlighted by Kuhn et al., who described broader inequities in access to sports participation and related healthcare resources [4]. Pilkington et al. further reported differences in treatment pathways and rehabilitation outcomes when com-paring elite and amateur athletes undergoing surgical repair for sportsman’s groin [5]. Particularly noteworthy is the study by Tavares et al., who directly compared elite and amateur rugby players and demonstrated significant differences in the usage and perceived effectiveness of recovery and rehabilitation modalities, with elite athletes having access to more advanced options [6]. Professional athletes are therefore regularly involved in comprehensive, multidisciplinary rehabilitation programmes with medical teams, while amateurs often have limited access to these care options, which may adversely affect their recovery. This disparity is further amplified by the lack of convergent injury-prevention strategies, psychological support, and specialised medical care [7,8]. A supportive environment, including coaches and trainers who understand the rehabilitation process, is common in elite sports. This support can facilitate adherence to rehabilitation protocols and improve recovery outcomes [9]. Nevertheless, despite significant research and technological innovation, many major questions remain unresolved, especially regarding the effective rehabilitation and long-term management of injured athletes, as well as return-to-play criteria. The authors also emphasised the need for a holistic approach [10,11].

Technological progress over the past few years has increased the potential of personalised wearable sensors, artificial intelligence (AI), virtual reality (VR), and neurocognitive training, in addition to improving real-time monitoring. These developments may improve the rehabilitation process and could foster more effective rehabilitation patterns [12,13]. At the same time, there is an increasing focus on biopsychosocial models, which assume that recovery from injury is shaped not only by physical factors but also by psychological readiness and social support structures. This approach enables more precise identification of risk factors and contributes to more effective prevention and rehabilitation strategies, which ultimately improves the outcomes of athletes’ return to full activity [14,15].

However, despite these promising directions, the current literature remains fragmented. The authors of several extensive studies emphasise that psychological and social dimensions are often ignored in clinical practice, while inconsistencies in methodology, small sample sizes, and the absence of high-quality trials continue to limit the strength of the evidence. It is also necessary to consider the inherent complexity and, consequently, the difficulty of evaluating the effects of rehabilitation interventions [16,17,18]. We often rely on the results of observational studies. In practice, creating high-quality randomised controlled trials (RCTs) is often a significant challenge. In addition, although many emerging technologies show potential, most remain in the early or intermediate stages of clinical readiness and lack verification in real-world settings. Digital technologies are expected to contribute significantly to rehabilitation. Newly introduced technologies, especially wearable devices and advanced machine learning algorithms, provide significant benefits to rehabilitation by improving monitoring, accuracy, and overall process efficiency. These technologies are changing the way athletes and other patients manage post-injury recovery. However, further research is essential to assess these risks, demonstrate a clear scientific advantage of digital methods over conventional methods, and ensure the broad acceptance and effectiveness of these new approaches in practice [19,20].

1.1. Objective

The aim of this review is to map and critically synthesise the evidence and gaps across three domains of sports injury rehabilitation: (1) emerging technologies (wearable sensors, artificial intelligence, virtual reality, and regenerative/orthobiologic modalities), (2) psychosocial/biopsychosocial approaches, and (3) implementation and methodological issues (standardisation, trial quality, and readiness for clinical translation).

1.2. Scope and Approach

This is a narrative review based on searches of PubMed, Scopus, and Web of Science (2018–2024). We prioritised systematic reviews, randomised and prospective clinical studies, and translational/implementation work relevant to the three domains; case reports and non-English sources were excluded. The synthesised themes (technology readiness and effectiveness, biopsychosocial use in practice, and implementation barriers) are presented in this article, and evidence gaps inform pragmatic research priorities and an implementation roadmap.

1.3. Methods

This work was conducted as a narrative review to map and critically synthesise evidence across three domains: (1) emerging technologies (wearable sensors, artificial intelligence, virtual reality, regenerative/orthobiologic modalities), (2) psychosocial/biopsychosocial approaches, and (3) implementation and methodological issues in sports injury rehabilitation.

1.4. Information Sources and Search Strategy

We searched PubMed, Scopus, and Web of Science for studies published between 1 January 2018 and 15 August 2025. Search strings combined terms for sports injuries and rehabilitation with intervention-specific keywords (e.g., wearables, artificial intelligence, machine learning, virtual reality, platelet-rich plasma [PRP], bone marrow aspirate concentrate [BMAC], stem cells, prolotherapy). Searches were complemented by citation chaining. Only English-language publications were included.

1.5. Eligibility Criteria

• Inclusion: Athletes or physically active populations; interventions involving emerging technologies, regenerative/orthobiologic therapies, or biopsychosocial approaches; study types including systematic reviews, randomised or quasi-randomised trials, prospective cohorts, and translational/implementation studies.
• Exclusion: Case reports (<10 participants), editorials, or opinion pieces without empirical data, non-English publications, and studies unrelated to rehabilitation outcomes.

1.6. Study Selection and Data Extraction

Titles and abstracts were screened, followed by full-text review of potentially eligible records. To mitigate bias in this single-author review, predefined eligibility criteria were applied consistently, and a 20% random sample of inclusion/exclusion decisions was re-checked after a cooling-off period. Extracted data included study design, population, intervention, comparators, outcomes, and key findings. For technology studies, additional fields (algorithm type, validation approach, implementation readiness) were captured.

1.7. Quality Appraisal

Methodological quality was assessed to contextualise the weight of evidence: RoB 2 for randomised trials, ROBINS-I for non-randomised studies, AMSTAR 2 for systematic reviews, and PROBAST for prediction modelling studies. For pilot VR/engagement trials, reporting completeness was considered. Quality assessments informed the narrative synthesis but did not determine study inclusion.

1.8. Synthesis Approach

Given the heterogeneity of populations, interventions, and outcomes, a narrative thematic synthesis was undertaken. The findings were mapped to the three prespecified domains (technology readiness/effectiveness, biopsychosocial practice, implementation/methodology).

1.9. Limitations

As this was a single-author narrative review, some risk of selection and extraction bias remains. To reduce this risk, predefined eligibility criteria, delayed re-checks of decisions, and triangulation with recent systematic reviews were applied.

All references included (n = 116) were classified according to the study design, see

Table 1. Summary of included references by study type.

Study Category	Count	Percentage of Total
RCT	1	0.9%
Observational (cohort/ case–control)	4	3.6%
Systematic review/ meta-analysis	24	21.4%
AI/ML prediction modelling study	2	1.8%
Other	81	72.3%

for details. Categories comprised randomised controlled trials (RCTs), observational cohort or case–control studies, systematic reviews and meta-analyses, and prediction modelling studies employing artificial intelligence or machine learning (AI/ML). The remaining references were assigned to an “other” category, including narrative reviews, expert opinions, conceptual frameworks, and methodological commentaries. This classification allowed us to map the balance of primary and secondary evidence and to identify gaps in high-quality trial data within the field.

To orient readers, Figure 1 synthesises the rehabilitation pathway across phases and illustrates how emerging technologies interface with the biopsychosocial (BPS) model. It highlights where wearables, AI, VR/AR, and orthobiologics contribute, from the immediate post-injury phase through late rehabilitation and return to sport/prevention.

We defined an analytic window of 2018–2024 for study identification and classification. A limited number of citations outside this window were retained for contextual purposes (e.g., foundational definitions, instrument validation, or seminal methodological work). These contextual citations were not eligible for synthesis and are reported separately as “references outside the window”, see

Table 2. Temporal coverage.

Earliest year	2000
Latest year	2025
Analytic window	2018–2024
References within 2018–2024	98
References outside window	14

.

2. The Technological Revolution: Wearables, AI, VR, and Beyond

The development of digital solutions in the field of sports rehabilitation has significantly increased, and the process of their diffusion is nonuniform. Modern technologies such as virtual reality (VR), wearable sensors, artificial intelligence (AI), and other immersive technologies are available to optimise both injury prevention and improve rehabilitation monitoring. These tools enable thorough examination of locomotion patterns, as well as enable the individualisation of treatment protocols with unprecedented accuracy, since they are able to generate non-stop, real-time biomechanical data. The benefit is that, although we cannot yet speak of explicit rehabilitation guidelines, several key takeaways and recommendations are highly relevant to the rehabilitation process [12,13]. Preliminary work indicates that AI systems paired with wearable sensors may identify risk-related movement patterns, and that personalised VR environments may enhance engagement. However, these signals derive from a few heterogeneous design- or framework-oriented studies without patient-level efficacy data or external validation, so they should be viewed as hypothesis-generating rather than confirmatory [21,22]. Specifically, Cucinella et al. [21] report a human-centred design process with experts—not a patient trial—and provide design recommendations rather than quantitative engagement outcomes. Aditi et al. [22] synthesise prior literature and propose an integration framework but do not report externally validated prediction metrics or clinical effectiveness, limiting generalisability to rehabilitation settings [21,22].

An important distinction exists between devices designed for professional and clinical use and those marketed for amateur or consumer athletes. Professional-grade wearables typically offer higher sensor accuracy, validated algorithms, and integration into clinical workflows, but their cost and technical demands limit widespread accessibility. In contrast, consumer-grade devices are relatively low-cost and widely available but often provide less accurate or less consistent data, which constrains their clinical utility and requires careful interpretation when used outside elite or research settings.

Wearable sensors enable continuous tracking of kinematics and early detection of biomechanical deviations that may signal elevated injury risk. However, their clinical uptake remains limited by heterogeneous data quality, non-standardised evaluation protocols, and the challenge of translating raw outputs into actionable rehabilitation decisions. When integrated with AI analytics (e.g., LSTM-based models), wearables can enhance real-time action recognition and responsiveness, but widespread adoption still requires external validation, interoperable data standards, and clear clinical pathways [13,19,23,24].

Applications in the AI sphere have produced impressive results, especially when defining the risk of injury and developing rehabilitation plans. Various experimental studies have shown that machine learning algorithms can perform better than conventional assessment methods in determining risk factors or making decisions related to return to play [25,26]. Nevertheless, clinical implementation is hindered by concerns about dataset reliability, ethical transparency, and generalisability across sporting populations [27,28,29,30].

Current evidence from machine learning applications in sports injury prediction remains preliminary, with most studies based on small pilot trials or retrospective datasets. Emerging ML architectures (e.g., LSTM-based models) offer promising methodological advances, but their clinical value will depend on rigorous external validation, cost-effectiveness analyses, and integration into standardised rehabilitation protocols.

Virtual reality (VR) and gamified environments can improve motor engagement, adherence, and cognitive stimulation—especially in postoperative or neurologically impaired athletes—when tailored to the individual’s motor/cognitive profile. However, most evidence comes from small pilot or laboratory studies, with limited longitudinal follow-up and fragmented implementation pathways; stronger clinical validation and standardisation remain priorities [20,31,32,33,34,35].

Although technological innovation undoubtedly represents one of the most dynamic areas of sports medical research, its effects are still constrained by fragmented implementation pathways and a lack of strong clinical verification. There is a clear need to go beyond studies and focus on structured, multidisciplinary trials that assess not only outcomes but also practical obstacles such as cost, access, and therapist training [17]. Although there is no denying technological potential, scaling up these innovations calls for interdisciplinary cooperation, including ethical governance, cost–benefit analysis, and clinical validation. Among the obstacles are the high implementation costs, differences in digital literacy, and a lack of standardised training and interoperability protocols [36,37]. Multimodal systems, such as AI-driven digital twins or AI-VR hybrids, potentially require coordinated approaches involving health economists, engineers, and clinicians.

Robotic rehabilitation is considered a promising and effective tool for injury rehabilitation, particularly for musculoskeletal and neurological conditions, offering advantages in terms of training intensity, monitoring, and cost efficiency. Although the current findings are encouraging, more robust research is needed to solidify its role and optimise its application in clinical settings [38].

The potential of three emergent technologies—artificial intelligence (AI), virtual reality (VR), and wearable systems—to support individualised sports rehabilitation is described in the above discussion. A comparison of several studies illustrating the way in which these technologies can be integrated is presented in

Table 3. Comparative summary of selected studies exploring AI and VR applications in rehabilitation.

Technology Used	Personalisation Level	Reported Outcomes	Integration Level	Study
AI + wearable sensors	High	Performance improvement, injury reduction	Strong sensor–AI linkage	[22]
LSTM, AI rehab planning	Moderate–High	Enhanced action recognition	Control systems	[23]
Logistic regression	Moderate	90% injury risk accuracy	Predictive modelling	[28]
VR neurorehab.	High	Enhanced motivation and engagement	Human-centred VR design	[21]
VR + robotic exoskeleton	Adaptive	Muscle response optimisation	VR–biomechanical feedback	[39]
AI robotics	High	Improved recovery rates	Integrated robotic control	[40]
AI for upper limb rehab	Moderate	Better outcomes than conventional rehab	Meta-analysis of RCTs	[27]
VR exergames	High	Usability-driven motivation, stroke rehab	Iterative patient-informed design	[34]
Intelligent devices + AI	High	Better adherence and quality scores	Sensor-driven planning	[41]
VR (systematic review)	Variable	Modest to good outcomes	Needs more standardisation	[35]

Abbreviations: AI: artificial intelligence; VR: virtual reality; LSTM: Long Short-Term Memory; RCT: randomised controlled trial.

. The studies were selected because they are relevant to personalised rehabilitation and include data on primary or secondary outcomes, as described in Section 1. This table shows the technological specifications of the study, the extent of personalisation, the result of rehabilitation, and the extent of system integration. The personalisation level refers to the extent to which interventions were performed according to dynamically customised interventions based on the physiological, biomechanical, or cognitive profile of the patient (Table 3).

The “Personalisation Level” column in Table 3 was assessed qualitatively based on the degree to which each study incorporated individualised elements into rehabilitation protocols. This categorisation is derived from the presence and depth of adaptive algorithms, sensor-driven feedback, and dynamic adjustment mechanisms tailored to each patient’s physiological or cognitive profile (see

Table 4. Explanation of personalisation level ratings.

Personalisation Level	Description	Examples from Studies
High	This technology provides real-time adaptation based on physiological, cognitive, or motor data. Personalisation is both dynamic and continuous.	[22]: AI-powered biofeedback sensors; [21]: VR adapted to cognitive profiles; [41]: Intelligent devices customising rehabilitation plans
Moderate	Initial personalisation is based on input data or assessments but limits real-time adaptation during therapy.	[23]: AI-based rehabilitation planning; [28]: Personalised injury risk prediction using logistic models
Variable	Mixed or inconsistent levels of personalisation often depend on the intervention design across multiple studies.	[35]: Systematic review covering VR with variable personalisation approaches
Low/N/A	Little or no adaptation to individual profiles; static rehabilitation content.	Not applicable to reviewed high-tech studies, but commonly seen in legacy VR or basic rehab protocols

).

3. Methodological and Practical Limitations

Despite the rapid expansion of sports rehabilitation research, the methodological foundations of several studies remain problematic. Several methodological shortcomings have hindered scientific progress and real-world translation. Chief among these are limited sample sizes, inconsistent study designs, poorly defined injury outcomes, and insufficient stakeholder involvement. These factors affect both the internal and external generalisability of the findings.

A recurring issue in several investigations is the small sample size used, which reduces the statistical power and limits the generalisability of the conclusions [18,42]. Over 30 studies within our search identified diverse methodological limitations, including poor reporting, inadequate sample sizes, lack of blinding, and inconsistent injury definitions. Small samples also prevent reliable subgroup analyses, particularly in terms of age, sex, or competitive level [43,44,45].

Other common challenges include the lack of standardisation of results and intervention protocols, as well as the lack of homogeneity of the reported results and the intervention procedures used in the studies. This variability makes comparisons across studies problematic and threatens the possibility of performing meta-analyses or formulating evidence-based clinical papers [17]. The classic evidence pyramid is still a rigorous method for evaluating complex rehabilitation interventions; however, little to no similar efforts have been made to question the effectiveness of individual rehabilitation modalities.

Furthermore, the implementation of these strategies is a practical barrier. Although effective in controlled settings, the prevention of several injuries fails to achieve traction in the real-world setting. The reasons for this discrepancy include poor stakeholder engagement, limited resources in community games, and a lack of integration between researchers and physicians. Together, these concerns hinder the adoption of evidence-based injury prevention practices in reality. Such impediments could be managed through long-term work to reach out to stakeholders, support injury prevention activities, and build strong links between the research and clinical fields [46,47]. The improvement in recovery procedures is tied to the further establishment of injury-monitoring systems that are more detailed and reliable. There must be systematic financing, greater surveillance coverage for amateurs as well as in multi-sport settings, utilisation of the latest technologies to record data, and consistent injury definitions through the engagement of healthcare experts. This will lead to a more precise understanding of risk profiles and, consequently, facilitate more efficient injury prevention and rehabilitation programmes. According to some authors [46], this may help implement “hybrid effectiveness–implementation” studies, which are designed to provide a more holistic and efficient approach to research, ensuring that effective interventions are not only identified but also successfully implemented and sustained in practical settings.

These gaps call for the use of hybrid effectiveness–implementation designs, which simultaneously assess both the clinical effect and practical feasibility of interventions. There is also increasing recognition of the need for causal inference frameworks, such as directed acyclic graphs (DAGs), to improve the interpretation of observational data and better understand injury mechanisms [16]. Implementation designs are needed to simultaneously assess the clinical efficacy and feasibility of the interventions.

Table 5 presents a thematic overview derived from the current literature and synthesises the most frequently reported methodological challenges in recent sports rehabilitation research. The table summarises five key domains, ranging from design and reporting limitations to external validity concerns, and provides illustrative examples from representative studies.

• Methodological Limitation Taxonomy. Across the sports rehabilitation literature, recurring weaknesses persist: incomplete/inconsistent reporting, small samples, lack of blinding, and non-standard definitions. Rehabilitation-specific challenges often amplify these issues. Clear research aims and appropriate (valid) statistical modelling are needed to reduce bias and improve reproducibility [43,44,45,48,49,50].
• Demographic Reporting Completeness. Although there has been some improvement, key demographic variables (race/ethnicity, socioeconomic status, age) are still frequently underreported; in paediatric/adolescent studies, age terminology is often inconsistent. This limits generalisability and complicates interpretation across populations [51,52,53,54].
• Intervention Description Adequacy. Therapeutic protocols are reported unevenly; sport-specific plans are often missing or insufficiently detailed, hindering replication and clinical translation. Standardised reporting tools/templates and full specification of intervention parameters are recommended [10,55,56,57,58].
• Study Design Rigour. Randomisation and the use of control groups are inconsistent, blinding is frequently absent, and small samples with short follow-ups reduce reliability. Several reviews recommend strengthening design standards and using suitable (e.g., multilevel) modelling to reflect data structure [44,45,48,59,60,61].
• External Validity Assessment. External validity is weakened by the limited description of participants and context, as well as data source biases. Mechanism-oriented interpretation and transparent context reporting are essential for transferability and real-world implementation [62,63,64,65,66].

This structured summary supports the identification of persistent research gaps and provides future directions for improving study quality and generalisability.

Table 5. Key methodological themes in sports rehabilitation research.

Category	Key Findings	Representative Studies
Methodological Limitation Taxonomy	Poor reporting, small samples, lack of blinding, inconsistent definitions; rehab-specific challenges; need for clear research aims and valid statistical models	[45]; [44]; [43]; [48]; [49]; [50]
Demographic Reporting Completeness	Underreporting of race, SES, and age persists; improvements noted but still insufficient; inconsistent age terminology in youth research	[53]; [51]; [52]
Intervention Description Adequacy	Inconsistent reporting of rehab protocols, limiting replication; sport-specific plans poorly documented; calls for standardised reporting tools	[55]; [56]; [10]; [57]; [58]
Study Design Rigour	Inconsistent randomisation and control group use; blinding often absent; reduced reliability due to small samples and short follow-ups; multilevel modelling recommended	[59]; [45]; [48]; [44]; [60]; [61]
External Validity Assessment	External validity weakened by poor reporting of participants and context; importance of mechanism-level understanding; data source biases affect generalisability	[62]; [63]; [64]; [65]; [66]

Table 5. Key methodological themes in sports rehabilitation research.

Category	Key Findings	Representative Studies
Methodological Limitation Taxonomy	Poor reporting, small samples, lack of blinding, inconsistent definitions; rehab-specific challenges; need for clear research aims and valid statistical models	[45]; [44]; [43]; [48]; [49]; [50]
Demographic Reporting Completeness	Underreporting of race, SES, and age persists; improvements noted but still insufficient; inconsistent age terminology in youth research	[53]; [51]; [52]
Intervention Description Adequacy	Inconsistent reporting of rehab protocols, limiting replication; sport-specific plans poorly documented; calls for standardised reporting tools	[55]; [56]; [10]; [57]; [58]
Study Design Rigour	Inconsistent randomisation and control group use; blinding often absent; reduced reliability due to small samples and short follow-ups; multilevel modelling recommended	[59]; [45]; [48]; [44]; [60]; [61]
External Validity Assessment	External validity weakened by poor reporting of participants and context; importance of mechanism-level understanding; data source biases affect generalisability	[62]; [63]; [64]; [65]; [66]

Table 6. Domains mapped to recent exemplars, precise shortcomings, and corrective methods.

Domain	Recent Exemplar(s)	Observed Shortcoming(s)	How to Fix (Scientific Method)
Methodological Limitation Taxonomy	[45]; [48]	Underpowered trials and incomplete statistical reporting; inconsistent definitions; aims not clearly linked to analysis. Allocation concealment/blinding frequently unclear.	Preregister studies; adhere to CONSORT; add a prespecified SAP (Statistical Analysis Plan). Conduct a priori power calculations; blind outcome assessors; prespecify. Apply GLMM/multilevel models with diagnostics.
Demographic Reporting Completeness	[51]; [53]	Underreporting of race/ ethnicity and SES; inconsistent age terminology; limited subgroup analyses.	Define a minimum demographic dataset (NIH/WHO categories) in protocol/CRFs; preregister fields. Plan stratified/interaction analyses by age/SES; adopt consistent age bands.
Intervention Description Adequacy	[5]; [58]	Incomplete description of intervention ingredients/progression; fidelity rarely reported; sport-specific content missing.	Use TIDieR (Template for Intervention Description and Replication) and CERT ( Consensus on Exercise Reporting Template). Publish full protocols/materials (e.g., OSF), including progression rules, equipment, and fidelity checklists; align outcomes to core sets where available.
Study Design Rigour	[59]; [48]	Randomisation/control groups inconsistently applied; assessor blinding absent; small samples; short follow-up; repeated measures analysed suboptimally.	Implement pragmatic/cluster RCTs with allocation concealment; blind assessors/analysts; use sham/attention controls when ethical. Conduct pretrial power calculation; extend follow-up period; analyse with ITT (intention to treat) and GLMM for repeated measures.
External Validity Assessment	[59]; [63]	Participants/setting/ providers underreported; mechanisms unclear; secondary-data biases limit transportability.	Use CONSORT-Pragmatic and PRECIS-2 in design/reporting; document context (setting, providers, resources). Embed MRC process evaluation (fidelity, mechanisms, context); conduct transportability/external validation analyses.

Abbreviations used in Table 6: CONSORT—Consolidated Standards of Reporting Trials; CONSORT-Pragmatic—CONSORT extension for pragmatic trials; SAP—Statistical Analysis Plan; GLMM—Generalised Linear Mixed Model; RCT(s)—randomised controlled trial(s); ITT—intention to treat; NIH—National Institutes of Health; WHO—World Health Organisation; CRFs—Case Report Forms; SES—socioeconomic status; TIDieR—Template for Intervention Description and Replication; CERT—Consensus on Exercise Reporting Template; OSF—Open Science Framework; PRECIS-2—Pragmatic–Explanatory Continuum Indicator Summary 2; MRC—Medical Research Council.

operationalises the domains summarised in Table 5 by linking each one to recent exemplars (DOIs provided), the concrete shortcomings observed (e.g., missing power calculations, incomplete intervention reporting, underreporting of race/SES, limited blinding, narrow sampling), and corrective methods grounded in recognised standards (e.g., preregistration with an explicit SAP; CONSORT and CONSORT-Pragmatic; TIDieR/CERT; ITT with appropriate GLMM; design for generalisability using PRECIS-2 and MRC process evaluation). This companion table is intended to move beyond general critique toward actionable guidance that supports replication, transparent reporting, and external validity in rehabilitation research.

We strongly emphasise that the “shortcomings” do not pertain to the studies cited in the tables in our manuscript; our intention is not to critique those authors’ work, but to convey the general deficiencies that these studies analyse and highlight.

Field-wide reporting completeness remains heterogeneous: most trials report age and sex, while race/ethnicity and socioeconomic status remain infrequently reported, with only modest between-study increases in recent years; para-athletes are rarely represented (e.g., Julian 2024 [51]; Talaski 2024 [53]).

4. Psychological and Biopsychosocial Dimensions: Still Neglected?

It is not a new finding that the psychological and social aspects of injury prevention and recovery play a role in defining treatment outcomes. The stress response is considered the most important risk factor for acute sports injuries. Earlier research also shows that psychological preparedness, motivational factors, phobias of any further injury, and social support play obscure but considerable roles in healing progress. Psychological readiness, motivation, fear of reinjury, and social support all affect rehabilitation results and return-to-sport decisions, yet these dimensions are rarely systematically involved in rehabilitation programmes [14,15].

Empirical studies continuously prove that using psychological tools (mindfulness-based stress reduction and cognitive behavioural therapy) can lead to better adherence rates, coping mechanisms, and control over emotions during the rehabilitation process [14,67]. However, such approaches are often on the periphery of clinical care, with standardised psychological evaluation tools or limited use of interdisciplinary care pathways. Based on the current evidence, it can be concluded that multidisciplinary or multimodal frameworks are generally not integrated into the routines of current psychological interventions used in clinical settings. In most cases, traditional, single-disciplinary psychological assessment tools and relatively narrowly defined care pathways are applied.

Biopsychosocial models are conceptualised to help practitioners relate the physical, psychological, and social aspects of the rehabilitation process. Despite this model’s theoretical soundness, its contribution to practice and empirical studies is limited, which, to a certain extent, is predetermined by the lack of empirically tested tools and the readiness of health service providers to administer such a comprehensive intervention due to the underdevelopment of their skills [15]. Integration is influenced by several factors, including sociological differences, personal beliefs, and mutual dynamics.

An intervention with a person-directed orientation that combines biological, psychological, and social components is required to optimise future rehabilitation. Accomplishing this prospect will require long-term cooperation between physiotherapists, psychologists, coaches, and physicians, in addition to interdisciplinary training and research strategies.

Table 7. Psychological and biopsychosocial factors in sports injury rehabilitation.

Intervention	Psychological Targets	Reported Benefits	Limitations	References
Cognitive Behavioural Therapy (CBT)	Fear of reinjury, anxiety, kinesiophobia, irrational beliefs	Enhances psychological readiness, reduces anxiety and irrational beliefs, supports mental resilience	Reduced effectiveness in telephone or education-only delivery formats	[68]; [69]; [70]; [71]; [72]; [73]; [74]; [75]
Biopsychosocial Integration	Psychological and social dimensions of recovery	Improved return-to-sport outcomes through interdisciplinary, individualised care	Incomplete integration in certain protocols; need for stronger holistic design	[68]; [73]; [76]; [15]; [77]; [78]; [71]; [79]
Readiness Assessment Tools	Psychological readiness, fear of reinjury, motivation	Validated tools (e.g., ACL-RSI, IPRRS, PRIA-RS) show strong psychometrics; support outcome prediction	Standardisation and refinement still needed for broader clinical adoption	[69]; [80]; [81]; [82]; [83]; [84]; [85]; [86]
Rehabilitation Adherence	Motivation, self-efficacy, fear avoidance, social support	Enhanced adherence through goal setting, psychological support, and tailored plans	Some studies showed no significant adherence differences across groups	[87]; [88]; [89]; [90]; [91]; [92]; [71]; [72]
Social Support Impact	Psychological resilience, motivation, confidence, anxiety	Support from family, coaches, and professionals improves recovery and return to sport	Communication gaps and lack of structured education for support networks noted	[70]; [83]; [93]; [87]; [94]; [76].

summarises the key evidence-based psychological interventions currently applied in sports medicine contexts to further support the integration of psychological and biopsychosocial dimensions into injury rehabilitation. Each intervention targeted specific psychological domains, such as anxiety, motivation, and emotional regulation, and demonstrated varying degrees of clinical benefit. However, implementation challenges remain, ranging from the lack of trained personnel and sport-specific validation to inconsistent integration within interdisciplinary care frameworks. This overview reinforces the need for the structured inclusion of psychological components in rehabilitation protocols and highlights the potential of targeted evidence-based strategies to improve both patient experience and functional outcomes.

Clinical takeaway: The following barriers are actionable and align with Table 7, which summarises evidence-based options (e.g., CBT-informed pathways, readiness tools, structured social support strategies) suitable for inclusion in standard rehabilitation programmes.

• Validated tools and standardisation. Limited availability and standardisation of sport-specific, validated psychological/biopsychosocial instruments and cut-offs (e.g., ACL-RSI, IPRRS, PRIA-RS) impede routine screening and monitoring [69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86].
• Provider capability and capacity. Insufficient training, time, and access to sport-specific psychological support, as well as weak referral pathways and uneven interdisciplinary workflows, reduce delivery fidelity [73,76,77,78].
• Workflow and infrastructure. The lack of integrated pathways, reimbursement/time pressures, and poor interoperability/digital capture of psychosocial outcomes hinder implementation at scale [17,36,37,46,47,63].
• Measurement and reporting gaps. Psychological outcomes are inconsistently measured and reported; demographic/context variables are often incomplete, limiting interpretation and equity [51,52,53,54,69,80,81,82,83,84,85,86].
• Evidence and external validity. Much of the literature is short-term or laboratory-based, with few pragmatic or hybrid effectiveness–implementation trials; generalisability remains uncertain [62,63,64,65,66].
• Cultural/attitudinal factors. Stigma, variable buy-in across teams, and limited structured education for support networks dampen adherence and sustained use [73,83,87,94].

Several validated instruments are available to operationalise these constructs. For example, the Anterior Cruciate Ligament–Return to Sport after Injury (ACL-RSI) scale is widely used to measure readiness and fear of reinjury, with values ≥ 65–70 generally considered acceptable for safe return to play [95]. The Injury–Psychological Readiness to Return to Sport (IPRRS) [96] and its adaptations (e.g., PRIA-RS) [84,96] provide additional structured assessments, while the Tampa Scale of Kinesiophobia (TSK) addresses maladaptive fear-avoidance beliefs. Despite these instruments’ potential, barriers to clinical implementation persist. These include a lack of clinician training in psychometric tools, limited availability of validated translations across languages, uncertainty about cut-off values in diverse sports populations, and low uptake when questionnaires are perceived as burdensome. Addressing these barriers by integrating psychological assessment into standard rehabilitation pathways, supported by digital platforms and interdisciplinary collaboration, could substantially improve patient-centred rehabilitation outcomes.

Table 8. Psychological factors influencing return to sport: validated instruments, cut-offs, and implementation barriers.

Psychological Factor	Validated Instrument (Cut-Offs)	Implementation Barriers	References
Readiness and Fear of Reinjury	ACL-RSI (≥65–70 indicates acceptable readiness)	Limited clinician familiarity; variability in cut-off use	[95]; [53]
General Psychological Readiness	IPRRS; PRIA-RS ( validated adaptations in multiple languages)	Translation gaps; not routinely integrated into clinical pathways	[96]; [81]
Social Support and Coping	IPRRS supplementary items; qualitative interview protocols	Low uptake in busy clinical settings; need for digital tools	[85]; [77]

ACL-RSI = Anterior Cruciate Ligament–Return to Sport after Injury scale; IPRRS = Injury–Psychological Readiness to Return to Sport scale; PRIA-RS = Psychological Readiness of Injured Athlete to Return to Sport questionnaire.

summarises key psychological domains relevant to rehabilitation progress and return-to-sport decision-making. For each domain, we list commonly used validated instruments (with cut-offs where available) and typical barriers to routine clinical implementation. The goal is to support consistent assessment, interpretation, and integration into care pathways.

5. Innovative Modalities and Regenerative Therapies

New treatment modalities for sports rehabilitation have attracted substantial attention in recent years, including orthobiologics, platelet-rich plasma (PRP), stem cell therapy, neuromuscular electrical stimulation, and neurocognitive training methodologies. Taken together, these modalities are often presented as powerful means of helping tissues heal, shortening recovery duration, and improving functional outcomes, especially when dealing with musculoskeletal injuries [97].

Although the initial results of the studies mentioned in the literature are positive, the adaptation of such interventions to clinical practice has limitations. Most existing studies rely on small-scale trials, case series, or short clinical follow-ups, which casts doubt on the generalisability of their results and the soundness of their conclusions [93]. Kassitinon et al. summarised several current trends in more detail: accelerated rehabilitation in some diseases, biofeedback integration, prolotherapy, orthobiologics, targeted corticosteroid injections, multimodal and opioid-sparing approaches, cannabidiol (CBD), telehealth, ultrasound, and a focus on specific populations and conditions.

Recent studies have focused on the mechanism of stem cell-based therapies for spinal cord injury (SCI), focusing on increasing plasticity, functional recovery, and neural network connectivity. However, the translation of these treatment benefits into generalised, clinically relevant results in human populations remains elusive and requires the production of stronger evidence. Moreover, significantly diverse guidelines on choosing patients and adopting treatment procedures among clinical studies make it impossible to develop uniform guidelines encompassing clinical conditions [94]. Of course, moral considerations and regulatory rules must be respected, such as the use of mesenchymal stem cells derived from adult tissues.

Meanwhile, AI-assisted physical therapy and smart devices have begun to influence the way physicians modify their design and rehabilitation protocols. Empirical studies have shown that these interventions have a significant impact on the rehabilitation process after sports injuries by providing tailored training programmes, targeted diagnostics, and continuous monitoring. Therefore, such technologies promote more effective recovery and demonstrate high efficacy rates, particularly for moderate-to-severe injuries that would otherwise be treated with mostly traditional and less effective therapies [41,95]. These devices require clinical verification and careful integration to ensure that they enhance rather than replace the therapeutic relationship.

Overall, while these innovative interventions offer exciting possibilities, their inclusion in mainstream rehabilitation is constrained by gaps in evidence, standardisation, and practical feasibility. Future research must prioritise large-scale, controlled studies with long-term outcome tracking while also addressing the logistical and ethical challenges associated with emerging therapies [96,98,99].

Platelet-rich plasma (PRP) therapy is widely used in sports medicine for tendon, ligament, and muscle injuries. It accelerates healing through the concentrated release of growth factors such as PDGF, TGF-β, and VEGF, which stimulate angiogenesis, fibroblast proliferation, and extracellular matrix remodelling. While systematic reviews report beneficial effects in pain reduction and functional outcomes, the results vary considerably depending on the preparation protocol and injury type [100,101,102,103,104,105,106]. Thus, although PRP holds clinical promise, heterogeneity in methodology limits definitive conclusions.

Bone marrow aspirate concentrate (BMAC) delivers a heterogeneous mix of mesenchymal stem cells, haematopoietic progenitors, and bioactive molecules that support tissue repair. It has been investigated particularly for cartilage, tendon, and osteochondral lesions. Clinical studies suggest improvements in pain and function, but outcomes remain inconsistent due to small cohorts, variable processing methods, and lack of randomised controlled trials [103,106,107]. Regulatory and standardisation challenges further complicate its broader adoption in routine sports rehabilitation.

Stem cell therapy, particularly mesenchymal stem cells derived from bone marrow or adipose tissue, represents an advanced regenerative strategy. These cells can differentiate into chondrocytes, myocytes, or osteoblasts, and they secrete paracrine factors that enhance repair. Preclinical models demonstrate encouraging tissue-regenerative effects, but clinical trials remain scarce and often methodologically limited [100]. Ethical issues, regulatory hurdles, and high costs also constrain translation to sports rehabilitation practice.

Prolotherapy involves injecting irritant solutions (commonly hypertonic dextrose) into injured ligaments, tendons, or joints to stimulate a healing response. The mechanism relies on local inflammation that triggers fibroblast activation and collagen deposition. Clinical evidence shows potential benefits for chronic tendinopathies and osteoarthritis, but trials are heterogeneous in design and often underpowered. As such, the therapy remains controversial and requires more high-quality, controlled studies to validate its role in sports rehabilitation [8].

Artificial intelligence-assisted rehabilitation robots represent a novel frontier in sports and neurological rehabilitation. These systems integrate robotic exoskeletons or end-effector devices with AI algorithms that adapt therapy intensity, monitor performance metrics in real time, and personalise exercise protocols. Pilot studies suggest enhanced patient engagement and functional improvements compared to conventional therapy, but widespread clinical implementation is hindered by high costs, technical complexity, and limited long-term outcome data. Further multicentre trials are required to establish cost-effectiveness and long-term efficacy.

This section provides an overview of alternative and advanced modalities in sports rehabilitation. To avoid conflating mechanistically distinct interventions, we present two complementary summaries:

Table 9. Regenerative/orthobiologic modalities.

Therapy	Mechanism	Evidence Level	References
PRP (Platelet-Rich Plasma)	Growth factors promoting healing and reducing inflammation	Moderate evidence from RCTs and meta-analyses; heterogeneous results	[103]; [102]; [104]; [105]
BMAC (Bone Marrow Aspirate Concentrate)	Stem/progenitor cells with paracrine and regenerative effects	Limited clinical RCTs; mainly observational studies	[103]; [107]; [106]
Stem Cells	Regenerative potential for musculoskeletal and neuronal repair	Early-phase studies; few controlled clinical trials	[108]; [100]
Prolotherapy	Injection of irritant solution to stimulate healing response	Sparse RCT data; mainly case series and observational reports	[104]; [8]

PRP = Platelet-rich plasma; BMAC = bone marrow aspirate concentrate; RCT = randomised controlled trial. Short comment for Table 9: Regenerative/orthobiologic approaches primarily target biological healing; while some meta-analytic signals are supportive, conclusions are tempered by protocol heterogeneity, small trials, and regulatory constraints—indicating selective, context-sensitive use alongside core conservative care.

synthesises regenerative/orthobiologic modalities (PRP, BMAC, stem cells, prolotherapy), whereas

Table 10. Digital and AI-Assisted rehabilitation modalities.

Technology	Application	Evidence Level	References
Wearables	Injury monitoring, load tracking, prevention analytics	Systematic scoping reviews; validation limited by device heterogeneity	[13]; [12]
Virtual Reality (VR)	Immersive rehab, engagement, motor learning	Growing evidence from RCTs and meta-analyses, but heterogeneous protocols	[35]; [32]
Robotics	Assisted training, gait/upper limb rehab, feedback systems	Systematic reviews support safety and efficacy; more comparative trials needed	[38]; [39]
AI Algorithms	Prediction models, personalisation of rehabilitation	Systematic reviews/meta-analyses show promise; external validation needed	[109]; [24]; [30]

Abbreviations: VR = Virtual Reality; AI = Artificial Intelligence; RCT = Randomised Controlled Trial. Short comment for Table 10: Digital and AI-assisted modalities enhance measurement, personalisation, and engagement, but their real-world value hinges on external validation, interoperable data standards, clinician training, and cost-conscious pathways—particularly beyond elite settings.

covers digital and AI-assisted rehabilitation technologies (wearables, VR, robotics, AI algorithms). This separation improves conceptual clarity and allows appraisal within appropriate evidence hierarchies and implementation pathways. It should be emphasised that much of the current evidence is derived from professional or elite contexts, and translation to amateur or community settings remains limited. Accordingly, these approaches are framed as potential adjuncts to standard rehabilitation rather than universally applicable solutions.

Regenerative modalities such as platelet-rich plasma (PRP), bone marrow aspirate concentrate (BMAC), stem cell therapy, and prolotherapy seek to enhance tissue repair through biological mechanisms [8,102,103,104,105,106]. Evidence for their use is heterogeneous, with some supportive meta-analyses but also variable protocols and limited long-term data. These interventions are mostly accessible in specialist or tertiary centres and often restricted by regulatory and cost considerations. Their use in amateur athletes should therefore be highly selective and considered only after conventional conservative care has been optimised.

Digital and AI-assisted modalities, in contrast, focus on measurement, personalisation, and patient engagement. Wearable devices, virtual reality (VR), rehabilitation robotics, and AI-driven predictive models represent rapidly evolving tools. Systematic reviews increasingly support their potential to improve monitoring, adherence, and outcomes. However, implementation challenges such as cost, data governance, clinician training, and patient burden remain barriers to widespread uptake, especially in non-elite populations. Low-cost consumer devices and simplified digital solutions may provide a pathway for gradual translation into broader practice.

Taken together, these two groups of modalities illustrate distinct but complementary innovation trajectories. While regenerative approaches aim to directly influence biological healing, digital technologies primarily enhance the precision, monitoring, and engagement aspects of rehabilitation. Their applicability to amateur athletes requires careful, context-sensitive adaptation, guided by both available evidence and practical feasibility.

6. From Bench to Field: Bridging Research and Real-World Practice

An ongoing problem in the rehabilitation of sports injuries is the research–practice disconnect. Although many of these interventions have proven to be effective and efficient, their implementation in real-life clinical or athletic practice is often limited by practical, organisational, and contextual factors. One example is the use of exercise-based sports injury prevention programmes (SIPPs). Despite significant evidence showing their effectiveness, their application in the field of public health and realisation of their proven advantages are currently inconsistent [46,110].

Injury prevention programmes have shown considerable success in reducing injury incidence, especially among adolescent athletes, through the realisation of effective injury prevention programmes, but are nevertheless underutilised at local and amateur levels. Some important barriers are the low level of stakeholder involvement, lack of resources and time, and low awareness or education of coaches and healthcare providers [47,111].

In addition, innovations such as digital rehabilitation platforms, sensor-enhanced feedback tools, and remote monitoring systems face scepticism about their purpose, cost-effectiveness, and real impact on the outcomes. Without meaningful end-user participation during development, these technologies often fail to meet the needs of clinicians or athletes. In cases where end users are not meaningfully involved in the developmental process, such technologies have a high propensity to fall short of the intended needs of clinicians or athletes [20].

A hybrid study design that combines clinical effectiveness with implementation results, such as feasibility, fidelity, and sustainability, is essential for translating research into practice. An implementation gap exists between research and practice and thus arises between the dissemination and adoption stages of an intervention development process. To address this dilemma, researchers and clinicians will have to work together from the start of intervention planning and implementation [17,46]. Engaging key stakeholders, including athletes, coaches, and multidisciplinary rehabilitation teams, can foster a shared sense of ownership and enhance the likelihood of adoption. Finally, closing the gap between the bench and field requires not only rigorous evidence but also practical knowledge and an understanding of context, communication, and constraints. Technologies and scientific novelties cannot permanently change things and situations when they are separate entities, but only when they are actively built into consistent and favourable systems.

7. Future Research Directions: A Roadmap

Future directions in the field of sports injury rehabilitation suggest a need for a more strategic, coordinated research agenda. Literature reviews point to a number of areas of high research priority that scholars should pay attention to as soon as possible, on the condition of further development of the field along clinically meaningful lines.

First, the main issue is the process of generating and validating individualised rehabilitation protocols. In most cases, existing guidelines place their expectations on blankets and non-adaptive guidelines, which fail to account for various differences in the type of injury, severity, psychological preparedness, and sensorimotor capability. Future research must thus take a step further to promote adaptive protocols using criteria that consider biomechanical and psychological data.

Second, wearable sensor technologies and AI-based devices must be clinically validated. These techniques are often introduced prior to a robust evaluation. Future randomised controlled trials and longitudinal cohort studies are critical for identifying the efficacy, cost-effectiveness, and practicality of different sporting groups.

Third, the psychological and biopsychological dimensions should be fully integrated into research and practice. This can involve the incorporation of mental health screening tools, the inclusion of psychological interventions in rehabilitation protocols, adherence to longitudinal studies, and an assessment of how these affect return-to-play results.

Fourth, new treatment methods (orthobiologics, nanotechnology, and stem cell solutions) require robust evaluation. The available evidence is not well organised and is heavily biased towards preclinical or observational reports, and few studies have clinical validity to be adopted on a larger scale.

Fifth, research on the methodology must be improved. Adopting causal inference frameworks, such as directed acyclic graphs (DAGs) and hybrid study designs, is critical in addressing biases, improving replicability, and linking mechanisms to outcomes.

Sixth, more attention should be given to underrepresented groups, including adolescent athletes and female competitive and paralympic athletes, whose injury patterns and rehabilitation course should be investigated specifically.

In conclusion, all six areas of interest can be seen as preceding a turning point in the development of sports injury rehabilitation science. It is necessary to approach each domain in a systematic manner and, in doing so, establish a research agenda that is not only relevant but also meaningful in a clinical sense and addresses the complexities of the modern sports environment [112].

While the standardisation of research settings is essential for generating robust and generalisable evidence, it is equally important to consider how psychosocial technologies and approaches can be translated into everyday practice. Potential strategies include the integration of brief validated tools (e.g., ACL-RSI, IPRRS) into community and amateur sports programmes, the provision of training for coaches and allied health professionals in recognising and addressing psychosocial risk factors, and the development of cost-effective digital applications to facilitate athlete self-monitoring. Partnerships between universities, sports federations, and local clubs may serve as a bridge for piloting such approaches outside academic centres. Moreover, sports associations could promote group-based workshops, peer-support interventions, and awareness campaigns that normalise attention to psychological readiness and social support. Such measures are likely to increase accessibility and ensure that the majority of the sporting community—non-professional athletes—can also benefit from psychosocial advances in rehabilitation.

By addressing these gaps, future research can better align the clinical requirements and real-world contexts, eventually leading to more effective, inclusive, and sustainable rehabilitation solutions.

8. Conclusions and Call to Action

Rehabilitation for sports injuries is critical. Scientific and technological developments provide unprecedented opportunities for personalised, data-driven, and interdisciplinary approaches. However, persistent gaps in evidence quality, methodological rigour, psychological integration, and real-world applications continue to limit their impact. This article highlights many major areas where progress is required and possible. Clinical validation and thoughtful implementation strategies should match the integration of emerging technologies such as AI, wearables, and VR. Similarly, biopsychosocial models should extend beyond theoretical support to include practical applications that integrate psychological and social dimensions into rehabilitation schemes.

Importantly, the future of effective rehabilitation lies in the collaboration between engineers, physicians, psychologists, data scientists, and athletes. Such cooperation is necessary not only to generate high-quality research but also to ensure that findings translate into practice in a way that is ethical, accessible, and sustainable. Therefore, the call for action is twofold. Researchers must design pragmatic, inclusive studies that answer not just “what works,” but “for whom, under what conditions, and how.” Clinicians and practitioners remain open to innovation, but they demand evidence and advocate patient-centred interdisciplinary care. By bridging the gaps between disciplines, between research and practice, and between innovation and implementation, we can move toward a future in which rehabilitation is not just reactive but truly transformative for every athlete at every level.