The Conspicuously Clandestine Epidemic: An evaluation of subconcussive head acceleration exposure in two seasons of elite men’s rugby union

Abstract

Rugby union is one of the world’s most popular team contact sports, with over 7.73 million players participating across 121 countries worldwide (Viviers et al., 2018). It is also a full-contact sport involving multiple collisions which brings an inherent risk of injury. It is estimated that elite-level rugby union players are exposed to approximately 11,000 head acceleration events across one season (Owens et al., 2021), with most injuries resulting from the main contact events including the tackle (24 - 58%), breakdown or ruck (6 - 17%), maul (12 - 16%), collisions (8 - 9%), and scrum (2 - 8%) (Fuller et al., 2007). Despite rugby union’s popularity, concussion injury rates are of growing concern (Hind et al., 2020; West et al., 2021; Yeomans et al., 2018) and there are additional concerns about the cumulative effects of repeated sub-concussions, which are inherently difficult to monitor (Caplan et al., 2016). Since 2016, an increasingly common method of monitoring total and peak exposure to head acceleration has been through the use of instrumented telemetry units (ITU) in the form of external sensors or mouthguards (Wu et al., 2016b). These ITUs usually provide triaxial accelerometer and gyroscope outputs that can be converted into g-force and radians per second respectively. This can be used to identify which events are resulting in a high magnitude of linear or angular acceleration whilst the player is participating in their match or training session, often identifying outwardly inconsequential contact as contributing to subconcussive loading.
The primary aim of this thesis was to quantify peak linear (PLA) and peak angular (PAA) head acceleration associated with different common contact events (tackle, ruck, scrum, maul and lineout) during professional rugby union matches. In addition, this thesis looked to outline variation in the magnitude of head acceleration events where contact event role, orientation or height were changed. Player position group and player position were also assessed to outline the varying experience of linear and angular head acceleration during contact events. Further investigation was conducted to summarise the differences in player experience of head acceleration events across two different tiers of English professional rugby union.
Seventy-one professional rugby union players, aged between 18 - 35 years old, participated in 22 matches whilst wearing an instrumented head acceleration monitoring unit (Protxx Inc., California, USA). A maximum of 23 players wore an ITU per match. The head acceleration events were reviewed using a custom-built video analysis software that allowed the “tagging” of HAEs to assign them a specific contact event type, orientation, specific player, player role, player position group, and player position. All video analysis was reviewed by a team of rugby union experts. Analysis of head acceleration events was conducted using a combination of Python Software Foundation, Python Language Reference, [Version: 3.11.]), RStudio (RStudio team, Boston, MA, USA) and SPSS 28.0 (IBM Statistics, NY, USA).
Across the two seasons, 20399 head acceleration events were collected, analysed and attributed to various contact events. The most common event was the tackle (frequency = 3774). The contact events with the highest median PLA were maul events (11.43g, IQR = 3.0), whereas median PAA was highest during tackle events (4528.46 rad/s2, IQR = 3847.35). Orientation and player role dictated the experience of head acceleration involved in each contact event. There were limited significant differences in head acceleration magnitude discovered between contact events. However, there were several statistically significant intra-event differences between player roles of the contact events. Premiership matches tended to have a higher median PLA (11.54, IQR = 0.85) and PAA (4765.38 rad/s2, IQR = 819.19). The Championship season had a higher per player per game contact event frequency (43.42). The player group that was exposed to the highest relative frequency of contact events was G3 (back row). G3 players were also exposed to the highest median PAA (4359.34 rad/s2, IQR = 4523.09) during and second highest median PLA (10.93g, IQR = 2.33) during contact events. G4 players (half-backs) had the lowest relative frequency of contact events accompanied by lowest median PLA (10.71g, IQR = 1.58) and the second lowest median PAA (3659.02 rad/s2, IQR = 3409.25). Flankers ranked highest when positions were compared using relative contact event frequency. Scrum-halves ranked last in terms of contact event frequency. Locks had the highest median PLA (10.95g, IQR = 2.45), whereas number eights had the highest median PAA (4654.94 rad/s2, IQR = 4440.10).
To mitigate for the high cumulative subconcussive load experienced by all professional rugby union players, it is important for legislators, coaches and managers to be aware of the events, positions and player roles that result in the accumulation of subconcussive head accelerations events. In order to manage a player’s exposure over a full career, it may be necessary to provide coaching interventions on contact event technique and apply limitations on playing time to adequately protect the future health of professional players.