Rugby union, which is the most popular worldwide team contact sport involving collision,¹has one of the highest levels of injury of all team sports.² Whilst there have been several epidemiological studies of injuries in rugby union at the amateur level,^3,4 the number of prospective studies amongst elite players is small. Targett⁵ presented a study of 25 professional players in a New Zealand Super 12 team over a 6 month period (15 games) and reported an incidence of 120 injuries per 1000 player-hours of exposure. Jakoet and Noakes⁶ described a study of the 16 teams in the 1995 Rugby World Cup in South Africa (55 games) and reported an incidence of 32 injuries per 1000 player-hours. Bathgate et al⁷ presented a study of the Australian Wallabies over the period 1994 to 2000 (91 games) and reported an incidence of 69 injuries per 1000 player-hours. Although two of the previous studies^5,7 reported numbers of training injuries, none of the studies reported the incidence of injuries in training. The differences in the reported incidence values highlight the importance of using a consistent methodology for epidemiological studies and, in particular, a consistent definition of injury.⁸ In the earlier studies, an injury was defined by Targett⁵ as something that prevented a player from taking part in two training sessions or caused a player to miss a match; by Jakoet and Noakes⁶ as a new injury that necessitated a player leaving the field of play; and by Bathgate et al⁷ as one that forced a player to leave the field of play or miss the next match. Noyes et al⁹ recommended that, if injuries only required the use of ice and bandaging, they should be excluded from epidemiological studies, whereas Luthje et al¹⁰ argued for all injuries requiring medical attention to be included because even minor injuries could affect an athlete’s long term physical and mental condition. An advantage of categorising all player complaints that require medical attention as an injury is that incidences of injury can then be compared across all sports. A disadvantage of this approach, however, is that the inclusion of large numbers of injuries, such as minor contusions and cuts, can place an overwhelming burden on physicians taking part in epidemiological studies in some contact sports.¹¹ The importance of reporting the incidence¹² and severity^13,14 of injuries in epidemiological studies has been emphasised in order to assess causal links between risk factors and injuries and to inform decisions on preventive and therapeutic interventions.

The aims of the research were to determine the aetiology, incidence, severity, and causes of injuries suffered by the England rugby union squad during their preparation for and participation in the 2003 Rugby World Cup in Australia.

METHOD

The method employed was a prospective study of the injuries and training practices of all players (forwards: n = 32; backs: n = 31) involved with the England squad during the 2002/2003 season and whilst they prepared for and participated in the 2003 Rugby World Cup. All players signed a consent form for their medical data to be collected and utilised in the study.

RESULTS

Table 1⇓ shows the average anthropometric measurements for the forwards and backs at the beginning and end of the study. There were significant differences over this period in body mass for forwards (p<0.001) and backs (p<0.001) and in skinfold measurement for forwards (p<0.01) but not for backs (p<0.20).

There were 145 injuries (match: 97; training: 48) reported during the study period (pre-World Cup: 67; training camp: 37; Rugby World Cup: 41) of which 22 were recurrent injuries. The average severity of all injuries was 14 days (new injuries: 12 days; recurrent injuries: 28 days). Table 2⇓ provides an overview of the distribution and severity of these injuries in terms of location and diagnosis and table 3⇓ lists the diagnoses of the 10 most frequent and the five most severe injuries. Only three concussions (average severity: 7 days) were recorded throughout the study period. There were no significant differences in the distributions of injury location (p<0.38) (table 4⇓), diagnosis (p<0.45) (table 5⇓), and activity (p<0.68) (table 6⇓) as a function of playing position (forwards: 67; backs: 78). There were also no significant differences in the distributions for injury location (p<0.09) (table 7⇓) and diagnosis (p<0.27) (table 8⇓) as a function of the activity at the time of injury (match: 97; training: 48). There were 47 “significant injuries” (match: 26; training: 21).

The players competed in 22 matches (pre-World Cup: 12; training camp: 3; Rugby World Cup: 7) during the 63 week period. The total exposure time was 8373 player-hours (match: 445; training: 7928) and the overall incidence of injury was 17 injuries/1000 player-hours (match: 218; training: 6.1 and forwards: 15; backs: 21). The overall incidence of “significant injuries” was 5.6 injuries/1000 player-hours (match: 58; training: 2.6). Table 9⇓ shows the incidence and average severity of injuries as a function of playing position, activity at the time of injury, and the phase of preparation for the World Cup. Table 10⇓ shows the incidence of injury as a function of injury severity. The major sources of match injuries (table 11⇓) were being tackled (23%) and rucking and mauling (16%), which equated to 50 and 35 injuries/1000 h, respectively.

The number and average duration of training sessions undertaken by forwards and backs are shown in table 12⇓, whilst the structure of the training programmes, as a function of the phase of the World Cup preparation and playing position, is shown in table 13⇓. A subjective assessment identified that the highest training intensity occurred during the training camp. Table 14⇓ presents the number and incidence of injuries as a function of each element of the training programme for forwards and backs. For forwards, 68% of endurance training involved off-feet exercise, such as rowing and cycling, whereas for backs, 95% involved running activities.

DISCUSSION

The issue of injury definition is contentious within epidemiological studies and difficulties can arise when making comparisons between studies if different definitions of injury are used.^18,19 Definitions similar to the one adopted in this study have been used previously in both individual studies, such as soccer,^15,20 volleyball,²¹ and basketball,²² and in large scale national audits of sports injuries, such as the Football Association in England for professional soccer,²³ the New Zealand Rugby Injury Audit and Performance Project for rugby union,³ and the NCAA Injury Surveillance System²⁴ in the United States. The definition of injury adopted within this study therefore enables data to be compared effectively across different studies and different sports.

The epidemiological studies reported by Bathgate et al⁷ and Jakoet and Noakes⁶ represent the closest studies to the present one in terms of sample population, whereas the study by Targett⁵ represents the closest in terms of injury definition. There were no incidence values reported for training injuries in these studies to compare with the incidence in the present study (6.1/1000 player-hours). The incidence of match injuries observed in this study (218/1000 player-hours) was nearly twice that reported by Targett,⁵ three times that reported by Bathgate et al,⁷ and seven times that reported by Jakoet and Noakes.⁶ The higher incidence of match injury observed in the present study can be explained in part by the 30% increase in the “ball in play” times reported for the 2003 Rugby World Cup compared to the 1995 Rugby World Cup²⁵: this longer time in play would increase the players’ risk exposure proportionately. The wider definition of injury adopted in the present study compared to previous studies and the efficient injury reporting regime implemented, as the team doctor was present at every training session and match, were also likely to contribute to the higher values reported here. A further contributory factor may be the increase in players’ body mass²⁶ and/or fat-free body mass²⁷ observed in the present study compared to previous measurements on British international rugby union players.

The distribution of injury diagnoses in the present study was different from those reported previously: this was caused in the main by the different definitions of injury. In previous studies,^5–,⁷ facial lacerations formed a major proportion (12–19%) of the reported injuries, as this type of injury resulted in players leaving the field of play for treatment. Although players were also required to leave the field of play in the present study for the treatment of lacerations under the International Rugby Board²⁸ “blood injury” law, these injuries did not, in general, prevent the players from continuing to play after treatment or from taking part in training sessions and matches: consequently, this type of injury was excluded from the present study. There were differences in the distribution of injury locations between the present and previous studies because lacerations to the head formed a major proportion of injuries in previous studies,^5–,⁷ but these were also largely excluded from the present study for the reason given above. The present study provided a detailed analysis of injury severity as a function of injury diagnosis and location and the playing position and activity at the time of injury. This level of detail was not available in the earlier studies. Two of the previous studies^5,7 categorised injuries as mild (⩽1 week absence), moderate (>1 to 3 weeks absence) and major (>3 weeks absence) and the results from the present study were similar to these previous results. Although it might have been anticipated that the more inclusive injury definition used in the present study might have increased the reporting of “minor” injuries compared with previous studies, this was not found to be the case. This may be attributable in part to the exclusion from the present study of minor laceration injuries that only resulted in the player leaving the field of play whilst the laceration was treated, whereas in previous studies these injuries were included.

It is not possible to compare the mechanisms of injury in the present study with previous studies as the level of detail provided in the earlier studies was limited: for example, two of the studies^5,7 did not differentiate between tackling and being tackled as injury mechanisms. The total proportion of injuries in the present study that occurred during the tackle phase (36%) was much lower than that reported previously^5–,⁷ (46–58%). Jakoet and Noakes,⁶ who did differentiate between tackling and being tackled, reported a similar proportion of injuries caused by being tackled (29% v 23%) but a much higher proportion of injuries caused by tackling (27% v 13%) than the present study. The lower total proportion of injuries caused within the tackle phase observed in the present study may, therefore, reflect a higher level of injury awareness by players during the tackling element.

From an injury management perspective, there are lessons that can be learnt from the results presented here in terms of the future development of training programmes. During the Pre–World Cup and the Rugby World Cup phases, 75–87% of training was aimed at the development of individual and team rugby skills in preparation for matches. However, during the training camp 52–55% of the training was aimed at improving the players’ physical condition. The success of the conditioning element of the training programme was indicated by the change in the players’ anthropometric measurements as body mass increased whilst skinfold measurements decreased resulting in a significant increase in the players’ fat-free body mass over the study period. Whilst the shift in focus from skills training to conditioning training had an impact on the incidence and severity of injuries observed during the training camp, only one player taking part in this phase was unavailable for selection for the World Cup squad through injury and this injury occurred during competition.

The overall incidence of injury during the study period was higher for backs than for forwards, although the average severity of injuries to the backs was lower than that for forwards. There was a consistent decline in the severity and a trend towards a lower incidence of injury amongst forwards for both competition and training over the three phases of the study. For backs, there was an increase in the incidence of injury over the three phases of the study for both training and competition and an increase in the severity of injuries during competition. There was, however, a large reduction in the average severity of injuries during competition and this resulted in the risk of absence due to injury (incidence×severity) being much lower during the Rugby World Cup phase. The incidence and severity of training injuries observed for forwards and backs during the training camp may reflect the accumulative effect of the volume and intensity of training undertaken at this time²⁹: a similar effect has been observed during international football training camps.³⁰ The overall reduction in injury severity during competition and the availability of all but one player for the World Cup suggested that the use of personalised conditioning training programmes for developing players’ strength and endurance was a successful injury management strategy.

Of particular importance from an injury management perspective was the demonstration that endurance running and contact training represented the highest risk activities. It is possible that the greater stature and body mass of forwards placed them at a higher risk during endurance running.³¹ However, by ensuring that a high proportion (68%) of the forward’s endurance training involved low risk off-feet endurance exercises, such as rowing and cycling, the overall level of risk within this element of conditioning training was reduced to the lowest levels possible. It may appear that the level of risk could have been reduced even more by further increases in the proportion of off-feet endurance training but taking this approach too far reduces the relevance of the training with respect to the specific demands of the game and would not have prepared the players for the harder ground conditions encountered during the World Cup. It has been suggested that training games involving skill based conditioning may provide a safer compromise,³² but this reduces the ability to control individual players’ endurance training. The development of individual and team skills involving player to player contact is a fundamental element of training for forwards and backs; however, the high incidence of injury associated with contact training, in particular for backs, indicated that a review of this aspect of training could be beneficial in reducing the incidence of injuries. Finally, whilst fitness testing has many benefits, it should not place players at an unduly high risk of injury and, therefore, the structure of test protocols should also be kept under review.