A Comparison between Australian Football League (afl) Injuries in Australian Indigenous versus Non-indigenous Players

It has previously been shown that being of aboriginal descent is a risk factor for hamstring injuries in Australian football. The aim of this study was to review the Australian Football League (AFL) injury database to determine whether there were any injuries where indigenous players had different relative risks to non-indigenous players. Analysis was conducted using data from the AFL injury database, which included data from 4,492 players over 21 years (1992–2012), covering 162,683 player-matches at AFL level, 91,098 matches at lower levels and 328,181 weeks (possible matches) of exposure. Compared to non-indigenous players, indigenous players had a significantly higher risk of hamstring injuries Conversely, indigenous players had a significantly lower risk of lumbar/thoracic spine injuries (RR 0.61, 95% CI 0.41–0.91), groin strains/osteitis pubis (RR 0.75, 95% CI 0.58–0.96) and Achilles tendon injuries (RR 0.32, 95% CI 0.12–0.86). The results for the above injuries were also significant in terms of games missed. There was no difference between overall risk of injury (RR 1.03, 95% CI 0.96–1.10) or missed games (RR 1.00, 95% CI 0.97–1.04). This suggests that indigenous AFL players have the same overall number of injuries and missed games, but a slightly different injury profile.


Introduction
There is a sizeable health gap that exists in Australia between indigenous (aboriginal) and non-indigenous people.Sadly, this gap also applies to injuries, with aboriginal people recording higher rates of transport injuries [1], higher rates of hospitalisation for burn injuries [2] and higher rates of intentional and unintentional injuries [3].In addition, aboriginal people often have worse outcomes from their injuries [4].The explanation for this gap is likely to be multi-factorial.Geography is likely to play a role as aboriginal people generally live in more remote areas [1], roads are worse and access to medical services is more limited.Socio-economic status is also likely to play a role, as it does in so many health outcomes.
In relation to sporting injuries, there is very little published research analysing indigenous injury rates.It has previously been shown that being of aboriginal descent is a risk factor for hamstring injury, independent of previous posterior thigh injury [5,6].Being of aboriginal descent has also been shown to be a risk factor for calf and quadriceps muscle injuries [6].The evidence is mixed in relation to hip and groin injuries.One study has shown that senior elite Aboriginal Australian rules football players suffered fewer hip and groin injuries compared to non-indigenous players [6].However, another study examined elite junior Australian rules football players and found they had a smaller range of passive hip internal rotation (and several other reduced hip and groin measures), suggesting that indigenous players may be predisposed to hip and groin injuries [7].There was a major confounder in this paper in that the non-indigenous players (from South Australia) were in pre-season training at the time, while the indigenous players (from the Northern Territory) were mid-season because in the Northern Territory they play in summer, rather than winter.The aim of this study was to look at current trends in injuries suffered by indigenous players in the Australian Football League (AFL).

Data
Data were obtained from the annual AFL injury survey.The methods of the annual AFL injury survey are now well established and have been previously described in detail [8][9][10].All teams now keep electronic records of injuries.Although teams use different systems, at the end of the season, injury data from each team are exported into a common system.
The standard AFL player contracts now include consent for players' injury records to be provided to researchers for the purposes of standard injury surveillance, on the condition that confidentiality is maintained over individual player injury records.The methods of the AFL injury survey have been approved by the AFL Research Board, which is the appropriate institutional review board for this type of study.The methods are observational only (non-interventional) and therefore this type of study (data analysis) is considered to be of negligible risk to participants by the Australian National Health and Medical Research Council [11].

Injury Definition and Coding
The definition of an injury is an "injury or medical condition which causes a player to miss a match" [8].It includes any injury or medical condition, sustained at training or during a match, provided it causes a player to miss a match.This definition was chosen to promote consistency between different clubs and seasons, and has the advantage of being more accurate and reliable, despite having some obvious limitations [12,13].Each club is required to complete weekly player movement reports which list the status of each player as either: (1) playing AFL (national league level) football; (2) playing football at a lower level (state/minor league level); (3) not playing football because of injury (either short or long term); or (4) not playing football for another reason (e.g., suspended or not selected at the AFL level and not participating in a lower level match that week).An injury is defined by this status, with the details then requested electronically by the injury surveillance coordinator at the end of the season for checking and analysis.These details include diagnosis, which is subsequently coded (using OSICS codes, version 9) [14,15].
For this study, data were used from 4,492 players over a 21 year period, covering a combined total of 162,683 player matches at AFL level, 91,098 matches at lower levels (e.g., state league) and a total of 328,181 weeks (possible matches) of exposure (1992-2012 seasons, inclusive).Indigenous status was self-identified and obtained from multiple sources [16][17][18].
Comparisons were made between indigenous and non-indigenous players across 33 different body categories, as well as overall.Relative risks and 95% confidence intervals were calculated in relation to both the risk of injury and the number of games missed.

Results
During the study period, 182 indigenous players were identified, which is about 4%.There was a bias towards a greater number of indigenous players in the latter years of this data collection period, as shown in Table 1.The average age (± SD) was 23.6 ± 3.6 years for indigenous players, compared to 23.6 ± 3.8 years for non-indigenous players.There were not enough data to compare height and weight.Indigenous players were listed for 7% of the "player seasons" over this time period which indicates that the average indigenous player had a slightly longer AFL career than the average non-indigenous player.During the study period, indigenous players played 13,156 matches at AFL level and 5,135 matches at lower levels (e.g., state league), compared to 149,527 matches at AFL level and 85,963 matches at lower level for non-indigenous players.This suggests that indigenous players were over-represented at AFL level (i.e., they were generally playing at a higher level).
Comparisons of the relative risks for indigenous and non-indigenous cohorts, including 95% confidence intervals, are presented below with respect to risk of injury (Table 2) and number of games missed (Table 3).As can be seen from the tables, comparing indigenous AFL players to non-indigenous players, of the 33 categories examined, indigenous players had significantly greater risks of hamstring injuries (RR 1.52, 95% CI 1.32-1.73)and calf strains (RR 1.30, 95% CI 1.00-1.69).Conversely, indigenous players had significantly lower risks of lumbar and thoracic spine injuries (RR 0.61, 95% CI 0.41-0.91),groin strains/osteitis pubis (RR 0.75, 95% CI 0.58-0.96)and Achilles tendon injuries (RR 0.32, 95% CI 0.12-0.86).The results for the injuries listed above were also significant in terms of games missed.There was no difference between overall injury risk (RR 1.03, 95% CI 0.96-1.10)or risk of games missed through injury (RR 1.00, 95% CI 0.97-1.04).
The average injury severity (i.e., the average number of matches missed per new injury) was very similar for indigenous (average of 3.69 games missed per injury) and non-indigenous (average of 3.78 missed games per injury) players.For hamstring injuries specifically, the injury severity was also very similar (3.28 games missed per hamstring injury for indigenous players versus 3.39 for non-indigenous players).

Discussion
Indigenous AFL players have significantly a higher risk of hamstring and calf strains, and a significantly lower risk of lumbar and thoracic spine injuries, groin strains and osteitis pubis and Achilles tendon injuries.Whilst player speed and running distance during games were not measured as part of the study, the injury patterns are suggestive of a sprint-athlete injury bias for indigenous players.Indigenous players may have a higher proportion of fast twitch muscle fibres, which is a commonly-accepted belief in the Australian sporting world although to our knowledge has not been specifically tested (e.g., by comparative muscle biopsies).Running faster makes you more susceptible to muscle strains, although it may protect against some overuse injuries as fatigue may limit overall workload.This is supported by previous research which has shown that sprinters get more hamstring injuries [19], whereas backache and hip problems were more common in middle-distance running [20].
In addition, the results that show no overall difference in injuries are interesting.Perhaps it could be inferred that when indigenous people are active, and have equal access to medical care, there is no injury gap.This also refutes any possible theory that indigenous people have bodies that are more frail or prone to injury.Unsurprisingly, it suggests that socio-economic status and access to medical services are more relevant in terms of addressing the injury gap more broadly.
In terms of primary prevention of indigenous injury in Australian football, implementing evidence-based programs shown to prevent hamstring and calf injuries (such as strengthening exercises, e.g., the Nordic hamstring exercise [21]) is recommended.
More broadly, looking at secondary prevention (i.e., reducing severity/morbidity), we consider that increasing access to medical services, in particular sports medicine services, is important.This is not just an issue for regional and remote areas.Specific sports medicine services for indigenous athletes are almost non-existent outside the elite football codes.

Limitations
We note that our study only compared indigenous and non-indigenous players.No other racial categories were used.However, there are no other major racial groups in the AFL, so this is unlikely to have been problematic.
Although there were more indigenous players in the cohort in the latter years of this study, the injury profile of the competition as a whole has not changed greatly over the 20 years so it is unlikely that the data analysis would be affected by the recent increase in indigenous player numbers in the competition.Interestingly, looking at the distribution of indigenous players in Table 1, it can be seen that indigenous players make up only 4% of the cohort, but 7% of the player seasons, meaning indigenous players tend to have longer playing careers.It is a minor potential confounder that indigenous players were, on average, more likely to be playing at AFL level than lower level although previous analysis has not found major variances in injury risk between the various match levels of play [6].
In addition, no multivariate analysis was conducted.This was primarily because player movement data were not available.If those data could be collected, this would be interesting to consider in future research, as comparisons of speed and distance could be made which would test our assumption/ hypothesis that the differing injury patterns may be due to different running patterns.

Conclusion
Indigenous AFL players have almost the same overall risk of injury and games missed through injury, but have a slightly different injury profile to non-indigenous players.They have significantly a higher risk of hamstring and calf strains, and a significantly lower risk of lumbar and thoracic spine injuries, groin strains and osteitis pubis and Achilles tendon injuries.The differing injury profile could be due to aboriginal players being 'speed' athletes, rather than endurance athletes.Primary prevention strategies could implement evidence based exercises to prevent muscle strains.Secondary prevention could focus on increasing access to sports medicine services.

Table 1 .
Indigenous players by season.
* indicates significance at p ≤ 0.05; Italics indicates significance at p ≤ 0.05 for both number of injuries and missed games; ~ACL, anterior cruciate ligament; MCL, medial collateral ligament; PCL, posterior cruciate ligament.

Table 3 .
Indigenous vs. non-indigenous relative risk of games missed through injury.