Abstract
Rugby union is a field-based, collision team sport comprising contact events. Injury surveillance studies have demonstrated that the incidence of match injuries is high, with concussion the most commonly occurring injury. However, there is growing concern and research focus on the long-term brain health outcomes associated with all repetitive head impacts during contact sport participation and not just those resulting in concussion. Recent wearable technology advancements have led to the development of instrumented mouthguards (iMGs), which are fitted with an accelerometer and gyroscope and have been validated and recommended for the on-field quantification of head acceleration events (HAEs).Concussion incidence and risk factors, and HAEs in elite rugby union matches have been well described. However, training remains unexplored, despite elite players spending the majority of their rugby time in training, and it being a relatively controllable and modifiable environment compared to matches. Understanding the risk of concussion and HAEs during elite rugby training during different periods in the season, and during different levels of contact training and activities, could support data-informed player-welfare policies and practices by governing bodies and practitioners, and objectively improve associations between HAEs and long-term brain health outcomes.
The aim of this thesis is therefore to understand the incidence and risk factors of concussion, and incidence of HAEs using iMGs, during elite rugby training. Chapter Two of this thesis reviews the literature to date and identifies the knowledge gap in the understanding of concussion and HAE risk in elite rugby union training. In Chapter Three, the implementation and application of iMGs in the elite rugby union training context are explored. This includes their validation, strengths, limitations, and ethical considerations, as well as their integration with other routinely collected data streams, the accuracy of head acceleration data sets, and an analysis of their magnitude. Chapter Four explores the suitability of training exposure data collected through injury surveillance for quantifying training exposure in rugby union epidemiology studies, by assessing the agreement between practitioner recorded data and video of corresponding training sessions. The discussion and analyses in Chapters Three and Four inform the experimental studies in Chapters Five, Six, and Seven. Chapter Five identifies that the incidence of concussion in training (0.23 per 1,000 hours) is very low compared to matches but shows it significantly increased between 2021 and 2024 (p<0.01) compared to the start of the study period. A generalised linear mixed model demonstrates that the pre-season (risk ratio (RR)=1.80 [95% CI: 1.24 to 2.60], p<0.01) and full-contact training (RR=2.05 [1.16 to 3.62], p<0.01) pose significantly increased risk of concussion, whilst match minutes played in the 28 days prior (RR=1.00 [1.00 to 1.00], p=0.11), and surprisingly history of concussion (RR=0.91 [0.49 to 1.65], p=0.75), do not. The findings of Chapter Five informed the investigation of HAEs during elite rugby training in Chapters Six and Seven, with their incidence described during the in-season and pre-season, respectively. The incidence of HAEs was highest during full-contact (men’s forwards=0.20 per player minute, men’s backs=0.16 per player minute) and repetition-based training activities (men’s forwards=0.25 per player minute, men’s backs=0.22 per player minute) in elite rugby training but this was where players spent a minority of their total weekly training time. The incidence HAEs in the pre-season (men’s forwards=0.11 per player minute, men’s backs=0.06 per player minute) was higher than that described during the in-season, with the incidence increasing for forward players during the middle (0.11 per player minute) and last three weeks (0.13 per player minute) of the pre-season period, coinciding with an increase in contact training volume aligned with periodisation. The magnitude of HAEs in training were very low compared to matches, with approximately only 5% exceeding 25 g and 1.5 Krad/s2 in peak linear and angular acceleration, respectively. Chapters Six and Seven also demonstrated the great inter-player variability in the absolute count of HAEs accumulated by individual players, relative to their positional average, during the same levels of contact training and activities.
The thesis concludes in Chapter Eight, with the original contribution to the literature summarised. Current injury surveillance training exposure data capture systems are accurate for calculating overall training exposure but lose accuracy when comparing levels of contact training. Whilst the incidence of training concussions is very low compared to matches, the pre-season period and full-contact training had a significantly higher level or risk compared to in-season and modified contact intensity training. The incidence of HAEs was highest during full-contact pre-season training. Repetition-based training activities observed a higher HAE incidence than game-based training activities during the in-season. There was high inter-player variability in training concussion and HAE risk. The data presented in this thesis should be used to benchmark population-level average HAE accumulation in training. The proposed methods for amalgamating iMGs with current routine data streams should be utilised to support a nuanced approach to balancing contact training time for preparing for competition whilst minimising unnecessary exposure to HAEs. This thesis fills the knowledge gap in the literature by demonstrating the risk factors associated with concussion, and the linked HAE exposure, during elite rugby union training, supporting a holistic understanding of overall HAE accumulation in the professional game.
| Date of Award | 24 Jun 2026 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Sponsors | Rugby Football Union & Premiership Rugby |
| Supervisor | Keith Stokes (Supervisor) & Dario Cazzola (Supervisor) |
Cite this
- Standard