Narrative review of youth football and sport injury research

We conducted a narrative review of literature relevant to youth football injury aetiology to provide a critical synthesis of existing literature and to specifically identify paradigmatic and methodological assumptions, research approaches, research methods and data analysis procedures. The focus on existing literature’s paradigmatic distinctions is an important first step in interdisciplinary work.3 According to Phoenix et al3 (p. 220), a ‘step back’ to understand underlying assumptions can advance knowledge and awareness of the respective research field as a way to ‘move beyond debates about right or wrong ways of approaching research’.

The first author has from January 2017 to May 2020 searched for literature on youth football injuries in the databases PubMed, Scopus, PsycINFO and Web of Science. Search terms included various combinations of ‘youth’, ‘soccer’, ‘football’, ‘injury’, ‘aetiology‘ and ‘risk’. Inclusion criteria were set to youth football players aged approximately 10–19 years and injury aetiology, injury prevention and/or injury risk factors. The search revealed a paucity of sociological youth football injury research. Thus, the search was broadened to sociological injury research in different sports and age groups, which identified research in biathlon, figure skating, rhythmic gymnastics, rugby and softball. To identify assumptions underlying paradigmatic distinctions, we predefined three analytic areas: body and injury perspective (state of the art regarding how an injury is explained and defined); paradigmatic assumptions (reality/truth/knowledge) and research approaches (methodology). Based on how these categories were approached in the literature included in the review, as well as discussions in the FIT project team and a presentation at a scientific conference,11 we placed the literature in the three paradigms dominant for youth sport injury research: positivism, postpositivism and interpretivism. Furthermore, five dominant disciplines of youth sport injury research were identified based on the reviewed literature; biomedicine (biomechanics, sport medicine and exercise physiology), sport psychology and sport sociology. Table 1 summarises key findings of the narrative review.

Reflections on distinctions of youth football and youth sport injury research

The narratively reviewed literature on youth football and youth sport injury research shows two paradigmatic distinctions. First, the literature is monoparadigmatic and tends to focus on one subdiscipline of sport science. Most of the literature is biomedical, sport psychological and sport sociological, and often based on either quantitative or qualitative research techniques. The French philosopher and sociologist Edgar Morin50 argues that by approaching research from single disciplines, complexity becomes invisible. The disintegration of complexity through monodisciplinarity is not due to the discipline in itself, but as Morin 50 (p. 2) emphasises, because ‘of the discipline as it is conceived, non-communicating with the other disciplines, closed to itself’. Monodisciplinarity in research reduces the phenomenon into separate parts, thus making the whole as well as interactions between parts and between parts and the whole, invisible.7 50 Second, and related to monodisciplinarity, there is no integration of research across the three paradigms.

These limitations can be clearly seen in our narratively reviewed literature included in table 1. Thus, current youth sport injury research can be seen to shield off—or in other words leave out—possibly important aspects, creating a picture that is not actually representative of the complex phenomenon under study (eg, injury aetiology).51 For example, biomedical and psychological disciplines, by mainly focusing on athletes’ physical bodies and minds, leave out their interaction with and in the larger sociocultural context the athletes participate in. Similarly, sociological research, by focusing on sociocultural context, often ignores how the physical body and mind interact with and influence athletes.

Researchers operating in these paradigms and disciplines can and do apply different research approaches and methods, thus, paradigms do not confine. Indeed, researchers have adopted approaches to be more inclusive, such as an ‘ecological’, ‘integrated’ or more ‘real-world’ approach to sport injuries.52–54 However, these approaches either miss to account for multiple and complex interactions between different ‘parts of the whole’,7 or do not attempt to integrate disciplines across paradigms or use multiple quantitative and qualitative research methods.

The complexity approach to sport injury research

The recognition that sport injury aetiology involves numerous interactions between various variables across multiple dimensions has led some researchers to explore the potential of a complex systems approach for understanding and researching sport injury aetiology.4 6–9 Although researchers understand and apply complexity to sport injury research differently, three elements characterise current discussions. First, a complexity approach embraces an understanding of context as open systems consisting of components that are actively connected through mostly non-linear relationships.4 7 8 A non-linear relationship between components in the system implies that A plus B does not necessarily equal C. For example, a weak muscle plus psychological stress does not necessarily result in an injury. Second, these non-linear relationships between the components can induce dramatic new effects giving rise to unexpected structures and events, also known as emergence.55 56 Emergence can, for example, be certain tendencies, powers or complex phenomena such as a sport injury. Emergence, or the injury and its aetiology is constantly evolving, and not strictly predictable.55 In other words, emergence is self-organising, meaning a complex system can transform over time, either growing or shrinking.55 Emergence should be accounted for when researching sport injury aetiology, for example, through longitudinal studies. Finally, and what Newell55 argues is typically overlooked in the understanding of complex systems, is the importance of ‘local knowledge’, or knowledge of specific parts of the system. An athlete can, for example, be stressed over an upcoming exam at school and is injured right before the exam. The athlete might believe exam stress caused the injury. A closer examination may, however, show that it was the way the coach had changed behaviour towards the athlete that affected the injury aetiology. When applying a complex systems approach to injury research, the focus moves from identifying single risk factors for injuries to recognising patterns of interaction among multilevel components, acknowledging that these components interact in unpredictable ways and may be moderated by a number of individual and contextual factors.4 6 8

Researchers are currently discussing how the implementation of complex systems thinking can advance sport injury research, specifically through statistical mathematical models and computer-based simulations guided by equations, rules and laws. Complex models4 and statistical procedures such as the Agent-Based Modelling (ABM) and Systems Dynamics (SD) modelling9 are recent examples. According to Bekker6 (p. 80), however, this development represents an “overemphasis on the epistemological question of how multifactorialism is accounted for in research and a corresponding underemphasis on the ontological considerations and assumptions we make about the world”, which reflects a dissonance in how complexity is understood and applied to sport injury research. The current argument for complexity assumes a reductionist positivist/postpositivist view where complexity is understood as emerging from the rule-based interactions of simple agents/elements and explored through agent-based modelling.56 Consequently, it tends to ignore that there is more to emergence than the product of interactions of agents; athletes are themselves complex systems, and more complex than agents in agent-based simulations.56 Thus, the current argument for complexity is limited in recognising the importance of understanding athletes’ interactions with and in a certain context; hence, it is ignoring the significance of incorporating sociocultural conditions that researchers from the interpretivist paradigm have identified as relevant. Bekker further argues that injury researchers’ focus should not be on methodological preferences, but rather on ‘understanding system goal behaviour using methodological pluralism’6 (p. 81). We agree. A complexity approach requires alternative research methodologies. One such approach, which we have found to have potential in the FIT project to account for the multiple dimensions of injury aetiology, is an interdisciplinary research process that includes scientific disciplines from different paradigms and integrates qualitative and quantitative research methods.

The potential of interdisciplinary sport injury research

Drawing on research by Klein and Newell,57 we understand interdisciplinarity as a research process aiming to address a topic that is too broad or complex to be dealt with adequately by a single discipline. Furthermore, an interdisciplinary research process integrates disciplinary perspectives through construction of a more comprehensive perspective.57 Both Newell55 and sports researchers such as Buekers et al58 argue that complex systems and phenomena are a necessary condition for interdisciplinary studies. The rationale for this argument lies in the way complex phenomena can be understood and studied. If an injury is to be understood as complex, then this would mean, as argued earlier, that the phenomenon is multidimensional. Seeing it from one single angle, such as from biomedicine or sociology, only, the phenomenon appears different than from another angle. Since the overall pattern of behaviour is non-linear and dynamic, an effective method for modelling such a phenomenon must offer insight into the separate parts as well as the complex pattern produced by their overall interactions.55 An interdisciplinary research process, which draws insights from relevant disciplines and integrates these insights into a more comprehensive understanding, is thus proposed to have potential to study sport injury aetiology.1

The benefits of interdisciplinarity have been pointed out in a recent Nature editorial.59 The editor specifically endorsed the incorporation of social sciences, cautioning that ‘[i]f social, economic and/or cultural factors are not included in the framing of the questions’ then ‘a great deal of creativity can be wasted’. Recently, sports sociologists Pringle and Falcous60 have also advocated a collaboration between the social and natural sciences. They argue that a collaboration between biomedical researchers and sport sociologists, which they refer to as methodological border crossing, could improve the political impact of research from the sociology of sport. Regarding sport injury research, Burwitz et al1 argued for the benefits of interdisciplinarity over two decades ago. Despite their call, however, very little empirical work has been conducted to this end. Reasons provided for this reservation have been related to various obstacles that prevent high quality, truly integrated interdisciplinary work, as well as practical difficulties relating to gaining funding and publishing.1 58 Philosophical obstacles relating to paradigmatic differences may also result in a lack of shared understandings, research approaches and methods,61–63 preventing researchers from adopting such an approach. Nevertheless, our research in the FIT project has demonstrated that interdisciplinarity has potential for sport injury research.

The interdisciplinary research process of the FIT project

The interdisciplinary research process of the FIT project consisted of five phases depicted in figure 1. The five-phase research process was developed based on key interdisciplinary research steps outlined by interdisciplinary researchers such as Alan Repko, Rick Szostak and William H. Newell.55 64 While presented sequentially, the five-phase research process is an iterative and non-linear process. At times it was necessary to move back to an earlier phase or forward towards later phases, which is common and reflects the complexity of the interdisciplinary research process.55 64

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

The interdisciplinary research process of the FIT project. The figure illustrates the five-phase interdisciplinary research process adopted for the study of youth football injury aetiology.10 The blue boxes present the main theme of each of the five phases of the process. The green boxes entail research actions. The interdisciplinary research process was not linear, but involved the researchers moving back to an earlier phase or forward towards later phases during the process.

In the first phase of the process, the FIT project research group came together from multiple scientific disciplines and agreed on researching injury aetiology in youth football using an interdisciplinary approach. In the second phase, the project team members decided which scientific disciplines to draw on to study the problem, a process that took place through meetings, discussions, reviewing literature and presentations of research within and outside of the project group. In phase III, we conducted research applying the following research methods to study youth football injury aetiology: questionnaire; measures of weight and height; injury diagnostics; clinical examinations; strength testing; running analyses; knee laxity measurements; examination of training protocols; field observations and interviews. Research methods were chosen based on knowledge gained from project team members’ disciplinary expertise, existing literature, internal meetings and discussions with experts outside of the research group. Additionally, phase III entailed recruitment of research participants.

In the final two phases, data were analysed through measurement-specific analyses (eg, three-dimensional kinematics to analyse movement-specific characteristics; isometric strength to analyse muscle weakness and/or muscle imbalance; content analysis of field notes and interview transcripts) and an analytic procedure was prepared that makes integrated analysis of measurement-specific analysis possible.

During the FIT project’s five-phase interdisciplinary research process, the research group encountered paradigmatic challenges typical of working interdisciplinarily.61 One example is sample size and recruitment. Where positivist paradigmatic distinctions require a statistically powerful sample size, the interpretivist paradigm demands possibilities for in-depth and detailed examination of one or a few cases. A further challenge was the lack of an integrated data analysis procedure that would allow us to integrate the measurement-specific analysis referred to above. As we could not locate a suitable analytical procedure, we needed to spend time and effort to develop and test such a procedure. Finally, we have found that presenting the FIT project research in multiple cases resulted in misunderstanding and scientific criticism. As a result, we had to place extra effort in considering the communication of the FIT project’s scientific rationale, methodology and results. Indeed, while we have not struggled to respect our different paradigmatic worldviews and demands, it has been challenging to find and communicate what interdisciplinary researchers like Repko and Szostak64 and Welch65 call the ‘space in between’ scientific paradigms and disciplines we have been trying to create. The key measure that helped us to move forward in the interdisciplinary research process was a negotiation process that entailed regular meetings.