Article Text
PDF
Abstract
Background The high injury rate associated with rugby union is primarily due to the tackle, and poor contact technique has been identified as a risk factor for injury. We aimed to determine whether the tackle technique proficiency scores were different in injurious tackles versus tackles that did not result in injury using real-match scenarios in high-level youth rugby union.
Methods Injury surveillance was conducted at the under-18 Craven Week tournaments (2011–2013). Tackle-related injury information was used to identify injury events in the match video footage and non-injury events were identified for the injured player cohort. Injury and non-injury events were scored for technique proficiency and Cohen's effect sizes were calculated and the Student t test (p<0.05) was performed to compare injury versus non-injury scores.
Results The overall mean score for front-on ball-carrier proficiency was 7.17±1.90 and 9.02±2.15 for injury and non-injury tackle events, respectively (effect size=moderate; p<0.05). The overall mean score for side/behind ball-carrier proficiency was 4.09±2.12 and 7.68±1.72 for injury and non-injury tackle events, respectively (effect size=large; p<0.01). The overall mean score for front-on tackler proficiency was 7.00±1.95 and 9.35±2.56 for injury and non-injury tackle events, respectively (effect size=moderate; p<0.05). The overall mean score for side/behind tackler proficiency was 5.47±1.60 and 8.14±1.75 for injury and non-injury tackle events, respectively (effect size=large; p<0.01).
Summary Higher overall mean and criterion-specific tackle-related technique scores were associated with a non-injury outcome. The ability to perform well during tackle events may decrease the risk of injury and may manifest in superior performance.
- Rugby
- Injury prevention
- Epidemiology
- Contact sports
- Adolescent
Statistics from Altmetric.com
Introduction
The tackle is the most common form of contact in rugby union.1 ,2 It is responsible for rugby having a higher incidence of overall injuries than other team sports.3–5 The tackle also contributes to the highest proportion of concussions and spinal cord injuries.6–8 Correct tackle technique is required for safe participation and success in rugby,9–11 and incorrect tackle technique is a major factor associated with injury.12 ,13 A better understanding of the contributing factors associated with tackle-related injury events is necessary to inform and develop appropriate and targeted injury prevention strategies.14
Video analysis may provide a detailed insight into the aetiology of injury events and has previously been used for injury epidemiology studies in basketball,15 ice hockey,16 handball,17 soccer,18 lacrosse19 and rugby union.13 ,20 ,21 The rugby-related studies reported on tackle characteristics associated with injury but did not evaluate the differences in tackle technique proficiency between injury and non-injury tackle events.
Gabbett and colleagues22–24 developed a list of technical criteria and used video footage to measure tackle technique proficiency in rugby league players. Although these criteria were reliable, they were developed using a senior player cohort in a controlled laboratory setting and did not include the analysis of actual match technique performance.
Therefore, we used real-match video footage of a cohort of high-level youth rugby union players to determine whether the tackle technique proficiency scores were different in tackles associated with injury in comparison to tackles that did not result in injury.
Methods
Overview
Injury surveillance was conducted at the South African Rugby Union (SARU) under-18 Craven Week youth rugby tournaments between 2011 and 2013, as described elsewhere.25 Video footage of the matches at each tournament was recorded.
Injury data
Details of each injury at the tournaments were recorded on the basis of the consensus statement for injury surveillance26 and stored in the SARU injury database. Player consent was acquired as described elsewhere.27 Video footage of the tackle-related injury events at the under-18 Craven Week tournament was accessed for analysis from the SARU video database. Authors were granted access to these databases by SARU and the UCT Human Research Ethics Committee (injury database HREC Ref: 438/2011; video database HREC Ref: R042/2013).
Tackle definition
A tackle event was defined as “any event where one or more tacklers (player or players making the tackle) attempt to stop or impede the ball-carrier (player carrying the ball) whether or not the ball-carrier was brought to ground”.9 ,21
Video analysis
On the basis of the injury collection data, the corresponding video footage for the tackle-related injury event was sourced. Video footage was analysed using Sports Code elite V.6.5.1, using an Apple iMac (Apple, USA). The analysis software allowed for control over the time lapse during each movement, and the recording and saving of each coded instance into a database. Depending on the player's role during the injury event, that is, ball-carrier or tackler, all of the player's previous non-injury events in the same role (prior to the injury event) were sourced as a control.
Technical proficiency criteria
Proficiency in tackling and ball-carrying technique was assessed retrospectively by the first author using standardised technical criteria. Separate proficiency criteria lists were created for front-on and side/behind tackles and ball-carries (the same technical criteria were used for side and behind events and were defined as tackles made from outside the ball-carrier's peripheral vision, as defined elsewhere).28 Each tackle was split into three phases; (1) precontact phase, (2) contact phase, and (3) postcontact phase. The lists were based on previous studies examining tackling proficiency22–24 and technique guidelines,12 and were critically assessed by a rugby research group comprising coaches, clinicians, sport scientists, trainers and administrators until a consensus was reached. The player was awarded either one point or zero depending on whether or not he achieved a particular criterion (all scores were calculated as arbitrary units). All injured tacklers and ball-carriers were assessed for both front-on and side/behind technique proficiency for non-injury events regardless of the direction of the tackle during the injury event.
Reliability
Five players were randomly selected using an online random number generator (http://www.random.org/). Each player was coded on two separate occasions, separated by at least 1 week.29 ,30 Intracoder reliability for the scoring of technical proficiency was assessed using the intraclass correlation coefficient (ICC)31 and the typical error of measurement (TEM). The ICC and TEM for tackle-related injury events were 0.99 and 0.63, respectively, and 0.98 and 0.58 for tackle-related events which did not cause injury. An external coder followed the same scoring procedure for the same five players and intercoder reliability was assessed using the same method. The ICC and TEM for injury events were 0.90 and 1.38, respectively, and were 0.96 and 1.31 for non-injury events.
Statistical analyses
Statistical analyses were performed using STATA V.11 and STATA V.12. A Shapiro-Wilk test confirmed that the data analysed were normally distributed. The Student t test was used to compare scores for each technical criterion on the list, the overall scores for each of the three phases of play and the overall score for tackler and ball-carrier proficiency between injury and non-injury events. Statistical significance was set at p<0.05. Mean scores and SDs were calculated for overall technical proficiency, for each of the three phases of play and for each technical criterion. Percentage scores, which indicated the total number of events for which the criteria score was satisfied, were also provided. Cohen's d effect sizes were calculated to determine the magnitude of the difference between injury and non-injury for overall technical proficiency, for each of the three phases, and for each technical criterion on the list for tacklers and ball-carriers. Effect sizes of<0.2, 0.2–0.6, 0.6–1.2, and 1.2–2 were considered trivial, small, moderate, and large, respectively.31 The tackle-related incidence rate was calculated for all injuries (including ‘time-loss injuries’) that occurred in the tackle, and for ‘time-loss’ injuries only (an injury that resulted in the player being absent for more than 1 match in a tournament, or more than 1 day of normal/planned recreational activities during or after the tournament. Injuries were confirmed as ‘time-loss’ injuries during the course of the tournament or via telephonic follow-up after the tournament. Weekly phone calls were made to assess the state of the injury until the player returned to practice). This calculation was based on the consensus statement for injury surveillance and involved dividing the number of injury events by the player match-exposure, and was reported per 1000 exposure hours with corresponding 95% CI.26
Results
Overview and incidence
Approximately 50% of all injuries (including ‘time-loss injuries’) recorded over the 3 years (2011–2013) occurred in the tackle (n=84 of 167). This translated into an injury incidence of 27 tackle-related injuries per 1000 exposure hours (95% CI 21 to 33). Furthermore, 21% of all injuries at these tournaments were tackle-related ‘time-loss’ injuries (n=35 of 167), which resulted in an injury incidence of 11 tackle-related ‘time-loss’ injuries per 1000 exposure hours (95% CI 8 to 15). The average (±SD) age, height, mass and body mass index of all injured players were 17.5±0.6 years, 181.6±8.8 cm, 88.3±13.1 kg and 26.7±3.2 kg/m², respectively.
Only 49 tackle-related injury events were available for analysis due to poor video quality or missing footage. A total of 48 players were injured in these 49 events (1 player was injured twice, first as a tackler and then as a ball-carrier; figure 1). On the basis of the 49 injury events, 23 were related to ball-carriers and 26 to tacklers. There were 248 non-injury events associated with the same sample of injured players that were matched to each player's role in the injury event (125 related to ball-carriers and 123 related to tacklers). A power analysis revealed that the sample size was sufficient to extrapolate the results to other playing cohorts (power=0.99).
Breakdown of the number of players injured, injury events, and player-matched non-injury events identified from the cohort of injured players. All events are categorised according to the player's role during the tackle and the direction of the tackle.
Performance outcomes
Approximately 31% of injurious tackles (n=8 of 26) and 72% of non-injury tackles (n=89 of 123) were ‘completed’ and ‘successful’, that is, the ball-carrier's progress was impeded by the tackler. In addition, 32% of non-injury ball-carries (n=40 of 125) resulted in an offload or a broken tackle, that is, ‘unsuccessful’ tackle, while only 9% of injurious ball-carries (n=2 of 23) resulted in an offload.
The ball-carrier's team was able to maintain possession of the ball in 78% of non-injury ball-carries that involved a breakdown (n=56 of 72). Only 50% of injurious ball-carries that involved a breakdown (n=9 of 18) resulted in the ball-carrier's team maintaining possession of the ball.
Tackle-related injury events
Tacklers were injured in 53% (n=26 of 49) of all tackle-related injury events and 47% (n=23 of 49) of injuries were to the ball-carrier (figure 1). Approximately 42% of the injuries to tacklers were during front-on tackle events (n=11 of 26) and 58% occurred during tackles from the side or behind (n=15 of 26). Of these, 54% involved only one tackler (n=14 of 26), 23% involved a simultaneous tackle by two or more tacklers (n=6 of 26), and 23% involved a sequential tackle (second tackler arriving after initial contact, n=6 of 26). Front-on tackle events were performed during 52% of injuries to ball-carriers (n=12 of 23), and 48% of injuries to ball-carriers occurred during tackles from the side or behind (n=11 of 23). Approximately 61% of these tackles involved only one tackler (n=14 of 23) with the remaining 39% involving a sequential tackle (n=9 of 23).
Non-injury tackle-related events
A total of 248 non-injury tackle events were identified from the same sample of injured players (figure 1). Of these, 123 events focused on the tackler and 125 events focused on the ball-carrier. Furthermore, 44% of events focusing on the tackler involved front-on tackles (n=54 of 123) while 56% of these events involved tackles from the side or behind (n=69 of 123). In addition, 53% of these tackles involved only one tackler (n=65 of 123) while 41% involved a sequential tackle (n=50 of 123) and only 6% were simultaneous tackles (n=8 of 123). Approximately 46% of events focusing on the ball-carrier technique involved front-on tackles (n=57 of 125) and the remaining 54% involved tackles from the side or behind (n=68 of 125). A sequential tackle took place in 51% of these tackles (n=64 of 125) and the remaining 49% of tackles involved only one tackler (n=61 of 125).
Ball-carrier front-on tackle proficiency
The overall mean score for front-on ball-carrier proficiency was 7.17±1.90 (ball-carrier proficiency=51%, n=12) and 9.02±2.15 (ball-carrier proficiency=64%, n=57) for injury and non-injury tackle events, respectively, for the injured ball-carriers (score out of a possible 14) (effect size=0.91, moderate; p<0.05) (figure 2). The overall mean score for postcontact technical criteria was higher for non-injury tackle events in comparison to injury events (effect size=1.52, large) (table 1).
Ball-carrier front-on proficiency results: injury versus non-injury (includes means±SD, % score for total number of events for which criteria score was satisfied, p values and effect sizes with interpretations)
Mean ball-carry technique proficiency scores for each front-on tackle injury (n=12) and non-injury (n=57) event, and for each side/behind tackle injury (n=11) and non-injury (n=68) event for the injured ball-carriers. Overall group means are also represented with SDs. *The overall group mean of a non-injury event is significantly greater than the overall group mean of an injury event.
Ball-carrier side/behind tackle proficiency
For tackles occurring from the side or behind, the overall mean score for ball-carrier proficiency was 4.09±2.12 (ball-carrier proficiency=36%, n=11) and 7.68±1.72 (ball-carrier proficiency=70%, n=68) for injury and non-injury tackle events, respectively, for the injured ball-carriers (score out of a possible 14) (effect size=1.86, large; p<0.01) (figure 2).
The mean score for ‘go to ground and present ball/offload’ (effect size=1.46, large; p<0.01) and the overall mean score for contact (effect size=1.95, large) and postcontact (effect size=3.94, large) technical criteria were higher for non-injury tackle events. Technique scores for ‘aware of tackler(s)/attunement’, ‘shifting the ball away from contact’, ‘head up and forward, eyes open’, ‘fending away from contact’, ‘explosiveness away from contact’, ‘ball protection’, ‘leg drive on contact’ and the overall mean score for precontact technical criteria all had a moderate effect size towards a non-injury outcome (table 2).
Ball-carrier side/behind proficiency results: injury versus non-injury (includes means±SD, % score for total number of events for which criteria score was satisfied, p values and effect sizes with interpretations)
Tackler front-on tackle proficiency
The overall mean score for front-on tackler proficiency was 7.00±1.95 (tackler proficiency=44%, n=11) and 9.35±2.56 (tackler proficiency=58%, n=54) for injury and non-injury tackle events, respectively, for the injured tacklers (score out of a possible 16) (effect size=1.03, moderate; p<0.05) (figure 3).
Mean tackle technique proficiency scores for each front-on tackle injury (n=11) and non-injury (n=54) event, and for each side/behind tackle injury (n=15) and non-injury (n=69) event for the injured tacklers. Overall group means are also represented with SDs. *The overall group mean of a non-injury event is significantly greater than the overall group mean of an injury event.
The score for ‘release ball-carrier and compete for possession’ (effect size=1.53, large; p<0.05) and the overall mean score for postcontact technical criteria (effect size=1.63, large) were both higher for non-injury tackle events. ‘Shortening steps’, ‘shoulder usage (drive into contact)’ and ‘arm usage (punch forward and wrap, ie, hit-and-stick)’ all had a moderate effect size towards a non-injury outcome (table 3).
Tackler front-on proficiency results: injury versus non-injury (includes means±SD, % score for total number of events for which criteria score was satisfied, p values and effect sizes with interpretations)
Tackler side/behind tackle proficiency
The overall mean score for side or behind tackler proficiency was 5.47±1.60 (tackler proficiency=46%, n=15) and 8.14±1.75 (tackler proficiency=68%, n=69) for injury and non-injury tackle events, respectively, for the injured tacklers (score out of a possible 12) (effect size=1.59, large; p<0.01) (figure 3).
The score for ‘pull ball-carrier with arms to ground’ (effect size=1.33, large; p<0.01) and the overall mean score for postcontact technical criteria (effect size=1.74, large) were both higher for non-injury tackle events. ‘Explosiveness on contact’, ‘contact target in centre of gravity’, ‘shoulder usage (drive into contact)’, ‘release ball-carrier and compete for possession’ and the overall mean score for contact technical criteria all had a moderate effect size towards a non-injury outcome (table 4).
Tackler side/behind proficiency results: injury versus non-injury (includes means±SD, % score for total number of events for which criteria score was satisfied, p values and effect sizes with interpretations)
Discussion
This was the first study to measure tackling and ball-carrying technique in real-match situations using a list of technical tackle criteria adapted from previous research on tackle technique proficiency22–24 and practical guidelines for safe and effective tackle technique promoted by BokSmart.12 We found that players had higher mean tackle and ball-carry technique proficiency scores during their non-injury events in comparison to the events in which they were injured (effect size moderate-to-large).
Precontact
During the precontact phase, tacklers were less likely to be injured when they shortened their steps prior to executing a tackle. Shortening steps and reducing speed prior to contact may assist in the tackler's decision-making. This may afford the tackler the opportunity to react to a sudden change of direction by the ball-carrier and allow him to orientate himself better to execute a safe and more effective tackle.12 Furthermore, reducing speed before contact may attenuate impact forces, which may also lower the potential for injury. Professional players have a higher risk of injury when entering tackle situations at a high speed.2 ,20
Ball-carriers were less likely to be injured if they were aware of the tackler's approach from the side or from behind. A greater proportion of injuries to ball-carriers may occur during tackles made from outside of their peripheral vision as they are unable to brace for the impending contact. A lack of player awareness contributes to tackle-related injuries in Australian rules football,7 professional rugby league32 and rugby union.13 ,28 All players, regardless of their role during the tackle, should be trained to develop peripheral awareness and to attune or adapt better to their surrounding environment. This will enable them to run into available space or prepare themselves for impending contact from various directions. An awareness of the impending contact may allow them to fend or drive themselves away from the tackler and maintain possession of the ball.
Contact
Our data indicate that tackling the ball-carrier in their centre of gravity (above the hips and below the sternum) reduces the chances of an injury outcome for tackles made from the side or behind the ball-carrier. Therefore, it is important for the tackler to orientate himself and track the ball-carrier prior to contact to make contact in this anatomical region. Making contact in this region is the safest for both tacklers and ball-carriers; tackling the ball-carrier below the waist may result in the tackler making contact with the hard and dangerous anatomical regions of the hips or rapidly moving legs.9 Similarly, tackles made above the line of the shoulders are dangerous for the ball-carriers and should be avoided as they are not permitted in rugby as stipulated by the laws of the game.2 ,9 ,20
Postcontact
The mean postcontact scores for front-on and side/behind tackles and ball-carries were higher for non-injury events and had a large effect size towards a non-injury outcome. This suggests that the ability to perform the necessary technical tackle-related criteria in the postcontact phase may reduce the likelihood of sustaining an injury. This finding could also mean that, by remaining injury-free during the precontact and contact phases of the tackle, the player is better able to execute the necessary technical criteria and remain uninjured during the postcontact phase, that is, an injury during precontact and contact may incapacitate the player and limit their ability to perform the necessary postcontact technical criteria.
Tackles that involved a shoulder drive after contact had an effect towards a non-injury outcome for tacklers during both front-on and side/behind tackles. This component of the technique is important for dominating contact situations and preventing the progression of the ball-carrier. This outcome may be dependent on the tackler's ability to execute precontact criteria such as shortening their steps to align their body for shoulder contact with the ball-carrier's centre of gravity. Active shoulder tackles increase in proportion with increasing age and playing level where players demonstrate experience in executing safe and effective tackle technique.13
The current study showed that going to ground and presenting the ball, for supporting players, was associated with a non-injury outcome for ball-carriers during the postcontact phase of the tackle. This finding may be due to the ball-carrier's ability to perform the required technical criteria better during the precontact and contact phases of the non-injury events. With this in mind, McIntosh et al13 observed that younger, less experienced players remained on their feet for a longer period and did not immediately go to ground after contact; therefore, this technical criterion may have important implications for avoiding penalisation, maintaining possession of the ball and for reducing injury risk.
Similarly, in this study, the criteria associated with a non-injury outcome for tacklers were pulling the ball-carrier to the ground, releasing the ball-carrier quickly, and competing for possession. The ability to perform these postcontact techniques and the correct preceding precontact and contact techniques during a tackle may increase the chances of regaining possession of the ball and minimising the probability of injury.
The ‘open’ nature of the tackle
The relationships between criteria in the different tackle phases should be appreciated as the tackle is a dynamic and open phase of play.28 ,33 The technical criteria performed in one of the three phases may be influenced by or may have an effect on the criteria performed in the other two phases. The failure of players to execute certain technical criteria before or during contact may have a negative effect on their postcontact performance and/or health outcome.
A limitation of this study is that it is challenging to determine the precise moment of injury during the tackle when using video analysis and it is not always possible to identify all tackle events due to missing or poor quality video footage. Future injury surveillance would benefit from high-quality video footage. In addition to this, although this study was sufficiently powered, the results of this study do not necessarily generalise to other rugby-playing cohorts. Senior professional players tend to exhibit superior tackling ability and make more frequent use of dominant, active shoulder tackles and miss fewer tackles in comparison to younger player groups.13 The tackle-related injury incidence was also found to vary across the playing levels. However, the specific tackle technique criteria detailed in this study is necessary for participation and may be extrapolated across all age groups and playing levels. Additional research is warranted to assess the use of this method in other player cohorts.
With this in mind, the technical criteria in this study could be used to assess technique proficiency and may identify key areas, across all three phases of the tackle, where players may need to modify their technique to improve performance outcome and reduce the risk of injury. These criteria could form part of a tackle technique training regime that is implemented by coaches to players from an early age.34
Conclusion
This is the first study to specifically measure tackling and ball-carrying technique proficiency using real-match video footage of high-level youth rugby union players. Players had higher mean tackle and ball-carry technique proficiency scores during their non-injury events in comparison to the events where they were injured. Having a greater overall technical proficiency and the ability to fulfil specific technical criteria were associated with a non-injury outcome. The findings from this study reaffirm current guidelines on safe and effective tackle technique, and may guide future coaching regimes and strategies for injury prevention.
What are the findings?
Using real-match video analysis, we found that higher mean tackling and ball-carrying technique proficiency may reduce the risk of injury during tackle events and may result in a positive performance outcome.
The ability to perform specific technical criteria during tackles and ball-carries may reduce the risk of injury.
The tackle is an open phase of play and correct technique performance should be emphasised during the precontact, contact and postcontact phases of the tackle.
The findings from this study may be used by coaches to assess and monitor individual tackle technique and may guide any modifications in technique.
How might it impact on clinical practice in the future?
Correct technique may guard rugby players from succumbing to moderate-to-severe injuries.
Rugby stakeholders, including sports physicians, may be better equipped to guide players injured during tackle events on the methods for preventing similar injuries in the future.
Acknowledgments
The authors thank the medical staff at all of these tournaments for their invaluable assistance with the injury data collection, and would like to thank Stephanus den Hollander for his role as an additional rater in the inter-rater analysis. They would also like to thank SARU's BokSmart National Rugby Safety Programme and Medical Department for commissioning the injury surveillance project and for granting access to the injury and video databases.
References
Footnotes
Twitter Follow Nicholas Burger at @it_was_burger, Michael Lambert at @MikeLambert01, James Brown at @jamesbrown06 and Sharief Hendricks at @Sharief_H
Contributors NB was granted access to analyse the data and was involved in conceptualising the manuscript; he also conducted the video and statistical analyses and wrote the initial drafts of the manuscript. NB, JCB and SH collected and entered the injury database data. All authors were involved in conceptualising and editing drafts of the paper.
Funding Funding for NB was provided by the National Research Foundation (Scarce Skills Development Fund Doctoral Scholarship).
Competing interests The injury data and video footage used in this study were provided by SARU and the BokSmart National Rugby Safety Programme. CR is the Senior Medical Manager and WV is the Senior Rugby Safety Manager at SARU. No payment was received from SARU for formulating this manuscript.
Ethics approval UCT Human Research Ethics Committee (injury database HREC Ref: 438/2011; video database HREC Ref: R042/2013).
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Owing to the sensitive nature of the information and appropriate medical ethics, access to a more detailed aspect of the available data set will be reviewed on request and on a discretionary basis. Sharing of only depersonalised and non-relatable data will be considered, once permission has been received from WV or CR (who can be contacted via email through the corresponding author—nicholas.burger@uct.ac.za).