Article Text
PDF
Abstract
Background The Whistler Sliding Centre (WSC) in British Columbia, Canada, has played host to many events including the 2010 Winter Olympics. This study was performed to better understand sliding sport incident (crash, coming off sled, etc) and injury prevalence and provide novel insights into the effect of slider experience and track-specific influences on injury risk and severity.
Methods Track documentation and medical records over 4 years (2007 track inception to 2011) were used to form 3 databases, including over 43 200 runs (all sliding disciplines). Statistics were generated relating incident and injury to start location, crash location and slider experience as well as to understand injury characteristics.
Results Overall injury rate was found to be 0.5%, with more severe injury occurring in <0.1% of the total number of runs. More frequent and severe injuries were observed at lower track locations. Of 2605 different sliders, 73.6% performed 1–29 runs down the track. Increased slider experience was generally found to reduce the frequency of injury. Lacerations, abrasions and contusions represented 52% of all injuries. A fatality represented the most severe injury on the track and was the result of track ejection.
Conclusions By investigating the influence of start location, incident location and slider experience on incident and injury frequency and severity, a better understanding has been achieved of the inherent risks involved in sliding sports. Incident monitoring, with particular focus on track ejection, should be an emphasis of sliding tracks.
- Luge
- Injury
- Retrospective
- Bobsleigh
- Skeleton
Statistics from Altmetric.com
Introduction
The Whistler Sliding Centre (WSC) track, built for the 2010 Olympic Winter Games in Vancouver, British Columbia, Canada, was homologated in March of 2008.1 The design includes 16 turns over 1.45 km of track with a vertical drop of 152 m from the bobsled start to the finish.2 The track includes seven different traditional start locations referred to as, from top to bottom, Bobsled, Men's Luge, Lady 1, Lady 2, Junior 1, Junior 2 and Tourist. The diagram in figure 1 illustrates the track configuration and relative start locations.
Whistler Sliding Centre track diagram.
Historically, bobsled, luge and skeleton athletes have been some of the most frequently injured in the Winter Olympic Games. Sliding sport athletes were identified to be the second most frequent medical clinic attendees of all the sports at the 2002 Winter Olympics in Salt Lake City, with 10.1% of all polyclinic visits being bobsled athletes, 8.7% being luge athletes and 0.7% being skeleton athletes, representing 19.6% of all polyclinic cases.3 Crim4 investigated the same event, using medical imaging as a proxy for injury severity, and found bobsled and luge to have a 0.8% and 2% risk of injury requiring imaging, respectively. This approach is limited as it provides no link to tissue-level injury as indicated in an objective medical diagnosis.
A study following the 2010 Olympics in Vancouver revealed 12% of all registered sliding sport athletes were injured, with the head, neck, upper spine and lower leg being the most prevalent regions of injury. Of particular, significance for this event was a luge fatality associated with a track ejection that occurred during training.5 These results stand in contrast to a study of sport-specific injuries in the 2012 Youth Winter Olympic Games in Innsbruck, Austria, that found sliding sports to rank among the lowest in relation to athlete injury rate with only 6% of athletes suffering injuries.6 This may suggest a track-specific influence on injury rate, different tracking methods or reflect safety measures applied in reaction to the fatality in 2010.
Despite high injury rates among sliding sport athletes in high profile sporting events, very few studies exist which attempt to identify specific injury prevalence and severity trends with relation to start location, incident location, and athlete experience level. Our objectives were to use statistical and biomechanical analyses of the WSC track over four winter seasons and to synthesise them to present descriptive statistics of risk factors including start location, incident location and slider experience. It was our aim to make our findings generalisable for the sliding sport communities, regardless of the specific track or sliding discipline.
Methods
We performed a retrospective injury analysis of the WSC track spanning the 2007/2008 and 2010/2011 winter seasons (inclusive) and did so as part of a larger ‘safety audit’ commissioned by Whistler Sport Legacies. Sports and disciplines in the study included: men's and women's singles luge, men's doubles luge, two-man and four-man bobsled, two-woman bobsled and men's and women's skeleton. Track documentation of gross usage of the track and incidents identified as potentially injurious (hereafter referred to as an incident) as well as videos of a number of incidents were provided by the WSC from track inception in December of 2007 to March of 2011. Information pertaining to the reviewed documentation is detailed in table 1. In addition, detailed medical records were obtained for athletes seen at the local medical clinics. Three databases were constructed: the run sheet database, the incident database and the medical record and abbreviated injury scale (AIS) database. Descriptive statistics were generated using all three databases. Relative risk analyses with 95% CIs were performed to examine the influence of track experience and start location on risk of injury using all track documentation and investigating all disciplines and start locations. It is important to note that incident was not synonymous with injury, in our use of the terms. For this analysis, injury rate is generally defined as the number of injuries divided by the number of runs.
Documentation and video included in the investigation
Track changes over the time period of this study include increasing the height of the wall in the exit of corner 16 on two occasions (an undocumented date between 20 February 2009 and 12 February 2010 as well as on 12 February 2010) and closure of the Men's start location on 12 February 2010.
Run Sheet Database
An entry was created for each run using the original run sheets which were electronically scanned. Analysis of this database allowed for an understanding of the correlation between start and incident location. Additionally, statistics pertaining to slider gender, age and experience were generated. It should be noted that slider experience, as used in this analysis, refers to the cumulative runs a slider performs on the WSC track. This database was also compared with the control tower logs and incident database as a quality check. Assumptions made for construction of this database are listed in table 2.
Assumptions made in the construction of the run sheet database
Incident Database
This database was compiled from all of the runs for which an incident was documented and was used to investigate incident location and injury prevalence at that location. Incidents which involved multiple sliders (ie, 2-man or 4-man bobsled or doubles luge) were only entered once in order to relate injury prevalence to specific track locations.
This database was interrogated to investigate the effect of start position on the severity of injuries, crash location and the descriptors of particular injury types. It included region (head, neck, torso, back and extremity) and nature (concussion, fracture, contusion, laceration and abrasion) of the injury. An initial assessment of injury severity was based on the assumption that a referral for subsequent medical treatment was indicative of a more severe injury. Table 3 lists the assumptions made for construction of this database.
Assumptions made in construction of the incident database
Medical Record and AIS Database
This database was compiled from medical records acquired from the Whistler Health Care Clinic (WHCC) and the Vancouver Organizing Committee (VANOC) Whistler Polyclinic (VANOC WPc) and enabled a more in-depth understanding of the injurious incidents that occurred on the WSC track. Use of the medical records was approved by the University of British Columbia Clinical Research Ethics Board (H11-02121). Appropriate ethical procedures were instituted to anonymise the records and maintain slider confidentiality. The documented injuries were used to determine injury severity using the AIS, maximum AIS (MAIS) and injury severity score (ISS).7–9 The AIS is an ‘anatomically’ based, consensus-derived global injury severity scoring system developed and administered by the Association for the Advancement of Automotive Medicine (AAAM) whereby severity is quantified on a scale from 1 (minor) to 6 (maximal and currently untreatable).7–9 MAIS takes the maximum AIS value for cases of multiple injuries.
The medical record and AIS databases were reviewed to determine injury severity as it relates to slider experience, sport and discipline, start location, incident location and the type of injury. Table 4 denotes the assumptions made for constructing this database.
Assumptions made in construction of the medical records database
Track ejection analysis
Incidents where near (sufficient vertical height to clear the track walls), partial (a portion of the slider's body exits the track) or full ejection from the track were investigated through available video footage and were analysed in detail. In these cases, we extracted the video frames showing this motion and evaluated the position of the slider's centre of gravity with respect to the track surfaces. This allowed us to evaluate the risk of ejection, or mechanism of ejection, at these locations. Medical reports of the sliders involved were scrutinised to understand the biomechanical mechanism and medical severity of the injuries sustained as a result of track ejection. The kinematic analysis focused on calculating change in velocity (delta V) of sliders that interacted with track structures as this measure is used in biomechanics literature to estimate injury risk.
Results
In total, over 43 200 runs spanning four winter seasons (41 months) were identified. Overall, the risk of an incident was 1.6% (total incidents over total number of runs), while the risk of sustaining an injury was 0.5% (total of any documented injury over total number of runs). The risk of sustaining an AIS 2+injury (moderate or greater injury) was less than 0.1%. Table 5 breaks down the total number of runs by gender, discipline and incidents.
Total number of runs by gender and discipline
Start location
The Run Sheet Database provided a means to evaluate the relationship between start location and incident prevalence. Figure 2 illustrates the per cent of total starts from a specific location resulting in an incident. The most significant incident prevalence was observed at the Lady's Start 2 with 3% of all starts resulting in an incident. Lady Start 1 and Men's start were also found to have a high risk of incident at 2.8% and 2.7%, respectively. A relative risk analysis of start location and injury prevalence showed that sliders starting from the Lower Ladies start or higher were 2.1 times more likely to be injured (statistically significant using a 95% CI 1.76 to 2.55).
Per cent of incidents per total runs for each track start location.
Experience
The Run Sheet Database in combination with the Incident Database allowed the relationship between slider experience and injury rate to be studied. Figure 3 shows the number of injured sliders (red) and the number of non-injured sliders (blue) stratified by total number of runs. Of the 2605 individual sliders, 86% took 1–29 runs, 76% of all sliders did 1–9 runs, while only 2% of sliders performed more than 200 runs.
Effect of experience on injury rate as a function of the sliders’ cumulative experience. Each data point for the injured sliders (light red bars) represents a specific injury and the cumulative runs they performed prior to the injury. If a single slider suffered multiple other injuries on subsequent runs down the track, ‘injured’ data points are added reflecting the increased number of cumulative runs.
Evaluation of injuries in the context of slider experience revealed that the highest injury rate was for sliders who had done between 30 and 59 runs. The injury rate dropped significantly once sliders reached 150 runs.
Evaluation of injury severity in the context of slider experience was done by incorporating the clinical records (figure 4). The injury severity information showed similar trends to the injury rate information. By investigating the relationship between injury prevalence and slider track experience through a relative risk analysis, sliders were found to be at a higher risk of injury with increased track experience by a factor of 1.9 (using a 95% CI 0.32 to 0.96). However, although not statistically significant, relative risk analysis suggested that AIS 2+injuries were more likely for less experienced sliders by a factor of 1.7 (95% CI 0.60 to 4.70).
Injury severity compared with slider experience for incidents with clinical documentation. Note: two injuries were coded as unclassified (AIS 9) due to a lack of specificity in the diagnosis. This included injuries suffered by slider B (figure 8) which contributed to his death. ‘Injurious incidents unclassified’ represents 100% for those bars without percentage labels. MAIS, maximum abbreviated injury scale.
Track location
Injury severity and frequency by track location were investigated using all three databases. Figure 5 depicts the severity of injuries at each corner, using the MAIS score and referral to the clinic as indicators. Corner 17 represents the outrun of the track. Overall, 710 incidents were recorded in the Run Sheet Database, of which 442 had details pertaining to injury and track location. Sixty-two sliders were referred to the clinic for further treatment, of which 46 medical records were available.
Injury severity and frequency (by % of total runs) versus track location. Corner 17 represents the track outrun.
Incidents in corners 13–17 represented more than 75% of the documented incidents and 69% of the sliders referred to the clinic. Of all the injuries scored, 74.5% were AIS 1, 23.5% were AIS 2 and 2% were unclassified. Nearly 60% of the incidents resulting in an MAIS 2 score occurred in or below corner 13. Incidents in corner 16 resulted in the most severe injuries. The only fatality on the track occurred at the exit of corner 16.
Of the incidents that occurred in corners 1 through 12, 43.6% were injurious with 20.2% of these being referred to the clinic. Incidents in corner 13 were found to result in injury 37.1% of the time, with 9.4% of all corner 13 incidents referred to clinic. Incidents in corner 16 were found to result in injury 62.8% of the time with a clinic referral rate of 23.3%. The outrun (corner 17) was also found to result in injurious incidents 46.4% of the time, with 21.4% of incidents resulting in referral to clinic. These injuries were generally due to side-to-side ricocheting as opposed to the slider being separated from the sled. Figure 6 depicts a diagram of the WSC track and uses circle size to represent injury frequency at each location.
Injury severity summary for each corner on the track. Circle size represents frequency. MAIS, maximum abbreviated injury scale.
Medical record review
The Medical Records and AIS Database was used to understand the nature of injuries sustained in the sliding sports on the WSC track. Figure 7 illustrates the most common injury types as identified through AIS classification while table 6 further classifies these documented injuries by sliding sport and discipline. Abrasions, contusions and lacerations represented the majority of injuries sustained on the track with soft tissue damage, fractures and head injuries also seen as significant.
Injury type by sliding sport and discipline for medically documented injuries
Injury type as determined through medical documentation provided by the Whistler Health Care Clinic and Vancouver Organising Committee Whistler Polyclinic (VANOC WPc).
Track ejection analysis
On review of all of the available documentation and footage, two events were identified for which the slider was deemed to have been partially or fully ejected from the track. Track video footage showed slider A (figure 8) being partially ejected; however, no supporting documentation of the incident or injuries sustained were available. Slider B (figure 8) was observed to be fully ejected and sustained fatal injuries as a result of interaction with structures outside of the track. Both instances occurred at the exit of corner 16 as the slider entered the final outrun and exhibited similar sled and slider kinematics. Both sleds contacted the ice wall (to slider's right), resulting in a significant vertical and leftward (rebound) motion. Figure 8 compares the two incidents from still frames taken from the track footage and shows that the low barrier on the slider's left was increased in height after slider A's incident. The barrier height was increased again following slider B's ejection. According to medical documentation, slider B's most severe injuries were caused by interaction with structures outside of the track. The kinematic analysis showed that in all cases where the slider interacted with track structures, but was contained within the track, the delta V was found to be substantially lower when compared with incidents where the slider experienced a full ejection.
Comparison of turn 16 exit kinematics for sliders A (left) and B (right) competing approximately a year apart.
Discussion
Start location
The retrospective analysis indicated a trend of increased incident and injury risk for the higher traditional start locations with the exception of the bobsled start (low risk from high start) and in-track starts from corner 11 (high risk from low start; figure 2). The increased risk with higher start locations was confirmed and found to be statistically significant in our relative risk analysis. It is expected that the reasons for the general trend include slider error propagating down the track and increased speed resulting in less reaction time and higher impact energy. It is expected that errors from the exit of one curve can lead to errors in the next curve with correction depending on slider ability and wall interaction. We speculate that the two exceptions can be explained as follows. At the Whistler track, the protocol is for new and recreational sliders to start from lower start locations and for more experienced sliders to start from a higher start location. Thus, in the case of the bobsled sliders, the injury rate is lower from a higher position because the sliders were in general more experienced. In contrast, the higher incident rate for starts from corner 11 may have occurred because these sliders have less experience both with the track and their discipline.
Experience
An investigation into the effect of slider experience on incident and injury rates revealed the peak risk beginning at 30–59 runs and this risk remaining relatively high to 150 cumulative runs and then decreasing. We speculate that, during the 30–150 run period, sliders progressively slid faster as their skills developed and started from higher track positions which resulted in a higher risk of both crashes and injury. As the slider accumulated more than 150 runs, slider experience and familiarity with the track resulted in reduced injury risk. The increased risk with exposure to more runs was confirmed and found to be statistically significant in the relative risk analysis. We also found a trend for less experienced sliders to suffer more severe injuries but this was not statistically significant. This suggests an interplay between experience and exposure where, after the slider's progressive introduction (which our results suggest was approximately 30 runs) the risk of injury increased with exposure to more runs. However, there was a point in the sliders’ experience that the risk of injury and the risk of severe injury decrease, although they are exposed to a higher number of runs.
Track location
Of note are the increased frequency and severity of injuries in corners 13 and higher. We hypothesise that this increased injury frequency and severity is indicative of increases in risk that occurred because slider speed increased with slider progress down the track, coupled with relatively complex turns in this section of the track.
Medical record review
The injury prevalence presented in this investigation is consistent with previous studies. Cummings et al9 found the risk of any injury for American luge athletes on the Lake Placid track to be 0.4% (per total number of runs) with risk of a more severe injury (loss of >1 day of training) to be 0.04%. The injury characteristics in that study were also generally comparable to our findings with contusions and strains representing 51% and 27% of the injuries suffered by luge sliders on the Lake Placid track, respectively.10 This was also consistent with the previous descriptive analysis of the sliding sport injuries in the 2010 Games where contusions and strains represented 30% and 27%, respectively, of the total number of injuries.5 It was observed that these lower severity injuries were difficult to track given many are untreated or treated by team physicians.
Track ejection analysis
Kinematic analysis and evaluation of videos of slider incidents revealed a drastic difference between incidents where sliders were contained within the track and incidents where sliders were ejected from the track. In the contained or non-ejection incidents, sliders were channelled down the track and their change in velocity and resulting impact forces were largely a function of the impact angle with the side wall or roof of the track. These contacts with the low friction surfaces predominately resulted in a change in direction of the sliders velocity vector with very little reduction of the sliders actual speed. In contrast, ejection incidents can result in sliders experiencing changes in velocity that approach their travel speed resulting in extremely high impact forces and in extreme risks for injury.
This analysis demonstrates the extreme importance of preventing all ejection incidents. Sliding accidents resulting from inexperienced sliders and unforeseen driver errors cannot be eliminated signifying the importance of track design in the prevention of slider ejection.
Strengths and limitations
By investigating the effect of slider experience and track-specific influences on the frequency and severity of incident and injury, novel insights into the associated risks of sliding sports has been achieved. Correlation of extensive track documentation and medical records allowed for unprecedented depth of analysis. Conclusions of this study can aid in improved safety for sliding sport participants through both track design and slider progression protocols. Although discipline-specific data were collected, the overarching goal is to present synthesised data to make generalised conclusions for the sliding community.
Several issues were encountered in performing this analysis, with incomplete, inconsistent or absent documentation being the most prevalent. Of the 710 incidents reported in the Run Sheets, only 63% were further detailed through Incident Reports and 25% through WSC Accident Reports. In some incidents, this made it difficult to confirm not only injury prevalence but severity and type as well. Medical reports acquired from the WHCC and VANOC WPc were only available for 74% of the athletes noted to have been referred for clinical follow-up, confounding our ability to assign AIS scores on those cases. Although some of the athletes referred to clinic may not have attended a medical facility due to the minor severity of the injury, some may have attended an outside clinic whose records were not available for review. Lack of specificity in the medical documentation led to the inability to classify several injuries on the AIS scale such as the head injury suffered by slider B (figure 8). Although this significant event is not represented in our severity data, minimal bias is present as injuries of all severities were found to be unclassified.
In order to maximise the benefit of this study, similar investigations at other tracks are necessary to build the strength of the trends identified. Such knowledge can lead to better protective equipment and sled design and can improve the design of future sliding tracks. Hubbard's10 2013 analysis provides an example of the benefit of case studies through direct application to track design and ice shaping.
Implications
Risk factors including start location, incident location and slider experience can be used to reduce injury risk for specific track's and across the sports of bobsled, skeleton and luge. Our results identify risk of an adverse event as it relates to start location (figure 2), slider experience (figures 3 and 4) and incident location (figure 5) and can be applied to slider development programmes. The development programmes may include cumulative run milestones and/or targeted performance goals before allowing athletes to progress to higher starts. More extensive experience from lower starts and education prior to graduating to higher start locations may decrease the risk of severe injury for athletes with little or no experience. For example, given that injury rate was observed to increase between 30 and 60 runs, our results suggest it maybe be beneficial to require new sliders to perform at least 60 runs at lower start locations prior to graduating to higher starts.
The strongest conclusion of this study is that every reasonable attempt be made to continuously monitor and evaluate track configuration and crash outcomes as they occur. Particular focus should be paid to mechanisms of potential track ejections. If identified, sliding should be immediately halted and corrective track modifications must be made before sliding resumes. In collecting and characterising incidents, track-specific modifications can be made to prevent similar incidents and the knowledge base can be incorporated into future track designs to avoid problematic features. In addition, protective equipment can be better tailored to address the most common and/or most dangerous scenarios.
What are the findings?
Incidents should be constantly monitored and reviewed to identify potential for track ejection or interaction with objects outside of the track.
Slider experience was generally found to be inversely proportional to the number of crashes and injuries.
The overall injury rate was found to be 0.5% which is comparable to a similar study conducted at the Lake Placid, New York track.
Track locations further down the track were observed to correlate with an increased frequency and severity of injury.
How might it impact on clinical practice in the future?
Reduced injury frequency and severity for sliding sport athletes in their initial training in the sport by implementing informed introduction protocols.
Improved track incident and injury monitoring procedures to identify and address problem areas to prevent future injury.
Safer track design through identification of features that may predispose sliders to increased injury risk.
Acknowledgments
The authors would like to acknowledge the Whistler Sliding Centre and Whistler Sport Legacies for funding this investigation and providing access to extensive track documentation. Furthermore, they would like to acknowledge Dr Jack Taunton and the Vancouver Organising Committee (VANOC) for providing access to the relevant medical records from the Whistler Polyclinic. They also thank Dr Kay Teschke for statistical insight and our colleagues at the Southern Alberta Institute of Technology and the Whistler Sliding Centre safety audit team, led by Alex Zahavich, for many fruitful discussions and help in conducting this study.
Footnotes
Competing interests None declared.
Ethics approval University of British Columbia Clinical Research Ethics Board. The approval certificate number is H11-02121.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement As this study was conducted as part of a greater Whistler Sliding Center safety audit, other data related to the injury analysis as well as other track data were made public in a report that can be found on the Whistler Sliding Center website or at the following link: http://www.whistlersportlegacies.com/about-us/reports/safety-audit