Thanks for submitting!

  • Jo Clubb

Understanding the Evolution of Injury Aetiology Models

Updated: Nov 14

There has been an increase in the focus on investigating the workload-injury relationship. Training load monitoring continues to be a major discussion point; in the literature, in the black hole of information and misinformation known as Twitter, as well as multiples articles on this very site.


However, we know training load alone is not the only factor for injury risk. Previous injury is consistently identified as a very important (if not the most important) risk factor for injury (Hagglund et al., 2006). Multivariate analyses demonstrate that injuries are not caused by training load or other risk factors in isolation but by the complex interactions of various risk factors.


Therefore, it is important to broaden our understanding of risk factors beyond training load alone. Appreciation of the causation of injuries in sport can be aided through understanding the history and evolution of injury aetiology models. In this article, the aim is to give an overview of some of the notable theoretical models of injury causation from the past three decades.


You can now also watch a brief video discussing these models (below).


Van Mechelen W, Hlobil H, Kemper HC. Incidence, severity, aetiology and prevention of sports injuries. Sports Medicine 1992: 82-99.

The journal site is here:

https://link.springer.com/article/10.2165%2F00007256-199214020-00002


The full text can be requested on ResearchGate here:

https://www.researchgate.net/publication/21646951_Incidence_Severity_Aetiology_and_Prevention_of_Sports_Injuries_A_Review_of_Concepts


Professor Willem van Mechelen’s 1992 paper is most frequently cited due to the model of injury prevention put forward. The “sequence of prevention” model involves a dynamic cycle through the following four steps:

  1. Establishing the extent of the sports injury problem.

  2. Establishing the aetiology and mechanism of sports injury.

  3. Introducing preventive measures.

  4. Assessing the effectiveness of these preventive measures by repeating step 1.

The stress-capacity model initially used in the Netherlands in relation to medicine (Ettema, 1973). Stress is determined by external factors and capacity by internal factors, and the two must be in balance to avoid injury.

  1. The stress-capacity model initially used in the Netherlands in relation to medicine (Ettema, 1973). Stress is determined by external factors and capacity by internal factors, and the two must be in balance to avoid injury.

  2. Van Mechelen presents a behaviour-oriented model adapted from works by Backx et al., 1990, Bol et al., 1991, and Kok and Bouter, 1990. It is proposed that the occurrence of sports injury is influenced by the interrelation between various risk factors, including the athlete’s physical and psychosocial factors, the environment’s physical and human factors, load, and personal equipment.

  3. Finally, van Dijk et al. (1990) updated the classic stress-capacity model into the stress-strain-capacity-model. This model acknowledges the relationship is time-based, that both stress and capacity are dynamic entities, and can also explain chronic injuries, which the previous two fail to do.

The sequence of prevention model has been a vital addition to the literature as a framework for research investigating the effectiveness of preventative measures. However, this paper is also a fascinating read regarding these early models of causation of sports injury.


Meeuwisse, WH. Assessing Causation in Sport Injury: A Multifactorial Model. Clinical Journal of Sport Medicine 1994: 166-170.


Available open access on the journal site here: https://journals.lww.com/cjsportsmed/Abstract/1994/07000/Assessing_Causation_in_Sport_Injury__A.4.aspx

Meeuwisse (1994) Figure 2. No copyright infringement is intended.

In 1994 Dr Willem Meeuwisse proposed a multifactorial model assessing causation of injuries in sport, in response to the limitations of a univariate approach. Meeuwisse develops Hennekens and Buring’s (1987) diagram of how multiple factors interact to produce disease, into a model for examining causation in athletic injury. The model discusses how intrinsic risk factors may determine the level of risk predisposed to the athlete, exposure to extrinsic risk factors may make a predisposed athlete susceptible and then an inciting event causes an injury mechanism.


This paper also includes a succinct and pertinent summary of assessing causation vs association, including errors that must be considered.


McIntosh AS. Risk compensation, motivation, injuries, and biomechanics in competitive sport. British Journal of Sports Medicine 2005; 39:2-3.


Available open access on the journal site here:

http://bjsm.bmj.com/content/39/1/2

McIntosh (2005) Figure 1. No copyright infringement is intended.

McIntosh also proposes a multifactorial model of injury causation but with a biomechanical focus on tissue properties and injury, rather than Meeuwisse’s epidemiological approach. The author states that injury results from a transfer of energy to the tissue, with the body’s response dependent on different factors depending on the tissue type. Thus, injury prevention interventions may be effective by reducing loads experienced by the body and/or by increasing load tolerance of the body.


Bahr R, Krosshaug T. Understanding injury mechanisms: a key component of preventing injuries in sport. British Journal of Sports Medicine 2005; 39:324-329.


Available open access on the journal site here:

http://bjsm.bmj.com/content/39/6/324.long

Bahr and Krosshaug blend aspects of Meeuwisse’s epidemiological model with McIntosh’s biomechanical model of injury causation. Internal and external risk factors need to be considered together at the time of injury, and not in isolation. Subsequently, how these factors can modify injury risk can be considered and ultimately incorporated into injury prevention programmes.


The authors propose the biomechanical properties of the inciting event should be comprehensively documented for the benefit of injury prevention research. Their proposed expanded definition of injury mechanism uses four categories of inciting events; playing situation, player/opponent behaviour, gross biomechanical description (whole body), and detailed biomechanical description (joint).


Meeuwisse WH, Tyreman H, Hagel B, Emery C. A dynamic model of etiology in sport injury: the recursive nature of risk and causation. Clinical Journal of Sport Medicine. 2007 May 1;17(3):215-9.


The journal site is here:

https://journals.lww.com/cjsportsmed/Abstract/2007/05000/A_Dynamic_Model_of_Etiology_in_Sport_Injury__The.11.aspx


The full text is currently available here.


In 2007, Meeuwisse and colleagues updated the multifactorial causation model to add a vital piece; the recursive nature of injuries. They acknowledge that injury risk is dynamic, rather than linear in nature; each repeat participation has the potential to alter risk. While intrinsic and extrinsic risk factors may change at different points in time, risk factors may also interact with each other to create joint interaction effects.


This model also acknowledges that there will be frequent occasions of athletic exposure without an inciting event that leads to injury. Thus not only is repeat participation considered, but also that the individual may cycle through the model via the differing pathways of potential adaptation from participation as well as recovery from injury.


Windt J, Gabbett TJ. How do training and competition workloads relate to injury? The workload—injury aetiology model. British Journal of Sports Medicine 2017; 51:428-435.


The journal site is here:

https://bjsm.bmj.com/content/51/21/1559


The full text can be requested on ResearchGate here.

Recently, Drs Johann Windt and Tim Gabbett recognised that despite the plethora of research demonstrating associations between training load and injury, to date, no aetiology model had explicitly incorporated workloads. They built on Dr. Meeuwisse’s 2007 recursive model to add the application of workload and attempt to explain how workloads influence injury risk.


Workloads are neither an internal (characteristic of the athlete) or an external risk factor (environmental aspect), and thus the authors present them as the vehicle in which athletes are exposed to external risk factors and potential inciting events. As well as this exposure, workload can influence subsequent risk via modifiable internal risk factors. Specifically, workloads can impart negative training effects by inducing fatigue and/or through positive adaptations that improve fitness.


Andersen MB, Williams JM. A model of stress and athletic injury – prediction and prevention. Journal of Sport & Exercise Psychology. 1988; 10(3):294-306.


The full text is available here:

https://pdfs.semanticscholar.org/4247/3a5952fcfae51fe74f3b28b5a3880d79f194.pdf


It would of course be remiss to discuss injury aetiology models without acknowledging those proposed from the sports psychology stream. While I’m sure many more exist that are not covered in this article, we should at least discuss Andersen and William’s stress and injury model.

In this model, the body’s stress response is driven by the athlete’s cognitive appraisal of the demand and their physiological and attentional responses to it. This response may determine the individual’s susceptibility to injury by affecting the ability to perceive environmental cues, via reduced visual field for instance, and/or disrupting coordination and flexibility, for example through increased muscle tension. Furthermore, this stress response can be intensified or mitigated by three factors; personality, stress history, and coping resources.


Bittencourt NFN, Meeuwisse WH, Mendonça LD, Nettel-Aguirre A, Ocarino JM, Fonseca ST. Complex systems approach for sports injuries: moving from risk factor identification to injury pattern recognition—narrative review and new concept. British Journal of Sports Medicine. 2016; 50:1309-14.


Available open access on the journal site here:

https://bjsm.bmj.com/content/50/21/1309

In this paper, the authors build on Meeuwisse’s dynamic, recursive model but argue a complex system approach is necessary to understand the nature of injury aetiology. Once again, literature on health and disease outcomes has a large influence as the source of the ‘web of determinants’ concept (Philippe and Mansi, 1998).


This complex systems approach urges us to try to understand the interactions between the risk factors rather than the risk factors in isolation. Risk factors have different weighting in terms of how they contribute to the system, along with different interactions with other risk factors. Therefore, we should look for risk profiles (the interactions between risks factors) rather than the separate risk factors. Crucially, the authors also acknowledge that risk factors may originate from different scales, such as biomechanical, psychological, and physiological.


Final Thoughts

I believe it is important to have an appreciation of the evolution of our knowledge. Sometimes we underestimate how long these discussions have gone on, presuming the current literature is ground-breaking, failing to acknowledge the prior research it builds upon. Risk factors for injury have been discussed in the literature since the 1970s, if not earlier. In 1984, Lysens and colleagues acknowledged only a fraction of risk factors had been identified. Despite being 35 years further on, it remains ignorant to assume we have uncovered all risk factors for injury.

While sports science may have become more reductionist in recent times, focusing in on training load as a/the determinant of injury, training load was actually only recently incorporated into published, theoretical models of injury aetiology. Although it does appear to be a risk factor for injury risk, it exists within a web of determinants that include many other risk factors.


Furthermore, we have moved away from univariate, linear approaches to injury causation, as we now start to consider dynamic, complex systems approaches. While it is more difficult to apply, it is important practitioners consider this complexity as they analyse load, response, and screening data.


Let’s finish with this quote from the late British statistician George E. P. Box:

“Essentially, all models are wrong, but some are useful.”