Documentation:FIB book/Design of Shoulder Pads to Mitigate Head Injuries Sustained due to Shoulder-to-Head Impacts in Ice Hockey

Background & Scope

Bodychecks in hockey can be high-impact injurious events.

Hockey is a fast-paced, contact sport in which collisions with other players are expected, if not encouraged. Bodychecking and body contact are two forms of player-to-opponent contact that hockey players may experience.

Bodychecking

Although bodychecking is banned in women's hockey, body contact, as shown here, is allowed and has the potential to cause injury.

Bodychecking (or checking) in hockey is a defensive tactic involving a deliberate “hit” of one’s body into a puck-carrying opponent, allowing a player to separate the opponent from the puck. Checks can occur at open ice, or against the boards and glass that surround a hockey rink. Speeds, areas of contact, and vulnerable organs may vary between these situations. Due to its deliberate nature and high likelihood of injury if practiced incorrectly, bodychecking is banned in Canadian minor hockey leagues for youth under the age of 13, and is banned universally in women’s hockey, including at professional and Olympic levels. Leagues in which bodychecking is allowed, such as the National Hockey League and 13+ Canadian minor hockey leagues, have recently introduced rules to ban checks in which the head is the primary point of contact. However, studies show that there has not been a significant reduction in concussions in leagues in which these rules have been added.

Body Contact

Body contact is a positional tactic used to legally block or impede the movement of the puck carrier. Body contact is an incidental form of contact that occurs due to the movement of the puck carrier into a skating lane that has been obscured by a defender’s body positioning. It is legal in all hockey leagues in Canada as it is considered a fundamental part of the game, and far less injurious than bodychecking. However, incidental contact can still result in injury, and head injuries caused by player-to-opponent contact still occur in hockey leagues in which bodychecking is banned.

Scope

This literature review will focus on head injuries sustained due to player-to-opponent contact, including both bodychecking and body contact. There has been extensive study on reducing head injury by changing the way in which the game is played through rule changes and refereeing, but as mentioned above, there is only so much that can be done in this regard. This review will target the issue from a different perspective, examining instead the design of hockey equipment to mitigate the risk of injury in shoulder-to-head impacts.

Epidemiology

An epidemiological study examined the injury rate per 10,000 athlete exposures (AEs) of high school hockey players in the United States participating in leagues that allow bodychecking. The overall injury rate was found to be 23.2 injuries per 10,000 AEs, with concussions forming the majority of injuries at 6.4 per 10,000 AEs, and bodychecking being credited as the most common mechanism of injury at 46 percent. A study by Wilcox et al. used instrumented helmets to collect data on the frequency and severity of different head impacts during games, finding that player-to-player contact accounted for 50% of all head impacts, a far greater proportion than any other mechanism. Other researchers used video analysis to classify head impacts, and corroborated the results of Wilcox et al., finding that collisions with other players or the boards accounted for 51.4% of all injuries, and that concussions were the most common injury at 18.6%. Another video-based study examined how concussions occurred in the National Hockey League, North America’s premier hockey league, using video analysis of plays that resulted in medically diagnosed concussions. It was determined that 88% of concussions resulted from player-to-player contact, and that 42% of these instances were caused by direct impact to the head initiated by the shoulder.

It is clear that there is a distinct correlation between player-to-opponent contact and concussion occurrence, and that shoulder-to-head impact is a prevalent cause of head injury. This data forms the motivation behind the topic for this literature review.

Injury Mechanisms

To better analyze the injuries sustained in hockey, it is important to first understand the circumstances under which impacts occur. A study was conducted in which 33 men’s university ice hockey games were observed, and data was collected on the circumstances of head impacts. In these 33 games, 75.9% of head impacts took place near or at the boards as opposed to open ice or near the net. Slightly more than half took place in the offensive zone of the player being impacted, with a third taking place in the defensive zone and the remainder in the neutral zone. Only 10.2% of head impacts occurred when the player had possession of the puck, with a relatively even split between players looking in the direction of impact and looking away. This data is for all head impacts, half of which are from impacts directly from another player. The study also found that players receiving head impacts experienced an average of 12.1 impacts each over the 33 games, compared to players delivering the contact each causing an average of 3.8 impacts over the same period. This suggests that a small subset of players are typically the victims of player-to-player head impacts. Another study found that out of a sample size of 13 shoulder-to-head collisions, 77% involved the shoulder moving laterally towards the head, 62% involved the shoulder moving only in the horizontal plane, and for 75% of the collisions the player delivering the check was either at a standstill or skating slowly at the time of the impact.

Out of the player-player head impacts, hand-to-head impacts are twice as common as both elbow-to-head and shoulder-to-head impacts. Interestingly, even though shoulder-to-head impacts are just as common as elbow-to-head impacts and far less common than hand-to-head impacts, they account for twice as many concussions as hand and elbow impacts combined. To determine the cause of this discrepancy, a study was conducted in which hockey players were asked to check a modified Hybrid III dummy’s head with their shoulder, elbow, and hand as hard as was comfortably possible, and the accelerations in the ATD’s head were measured. It was found that the higher rate of concussions in shoulder-to-head collisions can be attributed to relatively long impact durations compared to impacts from hands and elbows, even though the peak accelerations were generally lower. This study also provides baseline data for the linear acceleration, rotational acceleration, and impact duration that can be expected in typical shoulder-to-head impacts. For shoulder impacts to the front of the head, the mean peak linear acceleration and corresponding time-to-peak, and mean peak rotational acceleration and corresponding time-to-peak were 11.18 g’s, 12.31 ms, 511.4 rad/s^2, and 12.88 ms respectively. For shoulder impacts to the rear of the head, the mean peak linear acceleration and corresponding time-to-peak, and mean peak rotational acceleration and corresponding time-to-peak were 9.91 g’s, 12.20 ms, 534.8 rad/s^2, and 10.53 ms respectively.

While the longer impact durations in shoulder-to-head collisions seem to be the cause for the higher percentage of concussions compared to hand and elbow impacts, the chance of sustaining a concussion also increases with larger magnitudes of linear and rotational head accelerations. A study was conducted in which various shoulder-to-head collisions from real hockey games, some resulting in concussions and some not, were reconstructed with a modified Hybrid III dummy and various impacting methods to determine the accelerations in each case. The data was then analyzed to determine what accelerations correspond to 50% and 80% risk of sustaining a concussion. The linear accelerations resulting in 50% and 80% risk of concussion were found to be 14 g’s and 33.5 g’s respectively. The rotational accelerations resulting in 50% and 80% risk of concussion were found to be 1.95 krad/s^2 and 3.9 krad/s^2 respectively. Comparing these results to the mean accelerations typical in shoulder-to-head impacts shows that the average collision of this type is just below the threshold for 50% risk of concussion. The typical linear and rotational acceleration upper limits were found to be 16.1 g’s and 946.7 rad/s^2 respectively, which surpasses the linear acceleration value for 50% risk. This clearly shows that hockey players experiencing typical shoulder-to-head collisions are at a significant risk of suffering a concussion.  Adding to the dangers of concussion in hockey, it was found that only 16% of head impacts from contact between two players resulted in a penalty.

Injury Prevention

In general, protective equipment for contact sports is designed to protect the player wearing it, not the one being contacted. This is the case with ice hockey shoulder pads; they feature shoulder caps which are typically made of hard plastic or rigid foam which spread out the force of the impact, reducing the risk of shoulder injury . However, these hard surfaces can pose a threat to the player being contacted as they concentrate the bodychecking player's entire force into one surface. Studies suggest that modifications to the design of these shoulder caps could help protect the player receiving the check . There are three main methods of reducing the force transmitted to the player being struck: an increase in the duration and distance of the impact, dissipating the energy transmitted from one player to another, and allowing for a direction change in force during impact (sliding instead of direct transmission).

A 2017 study by Virani et al. investigated the effectiveness of adding a layer of foam to the outer surface of the shoulder caps in reducing the severity of impacts to the head. Specifically, a 2cm layer of polyurethane foam was molded to the outside of the shoulder caps and fifteen players bodychecked a dummy in the head with and without the foam-padded shoulder pads. The results of the study were highly promising, showing a 25.0% reduction in the average peak linear head acceleration, and a 12.4% reduction in peak rotational head velocity for checks with the foam padding. In terms of head injury prevention, these results are significant because with the extra padding, none of the impacts resulted in a peak head acceleration over 60g, which has been established as the minimum threshold value for which concussions can occur. Without the foam padding, four of the fifteen players were able to deliver an impact above this threshold. Therefore, adding a foam layer to the outer surface of shoulder caps has potential to serve as a simple yet highly effective method of reducing head injury risk and severity in ice hockey.

Richards et al. analyzed the decrease in peak resultant linear head accelerations and HIC values due to shoulder pads in shoulder-to-head impacts. Their research focused on different types of shoulder pads: tethered, traditional, and integrated shoulder pads. To perform the study, one ATD was stationary and upright while another ATD was swung shoulder-first into the stationary ATD on a pendulum. Each shoulder pad type was tested 5 times to ensure accurate results. The authors discovered that all three shoulder pads styles reduced the peak resultant linear head acceleration and HIC values of the ATD by 35-56% and 48-59%, respectively, when compared to no shoulder pads. Furthermore, a reduction of 18-21% was observed in linear head accelerations using the integrated shoulder pads versus the other two styles. While the angular accelerations experienced by the stationary ATD did not approach the threshold for severe brain injury, they did consistently approach concussion thresholds.

Please refer to Figure 1 in Richards et al.'s study for images of the three different styles of hockey shoulder pads

Virani et al. linear head acceleration values

Richards et al. linear head acceleration values

As seen in the figures above, both studies discovered that tethered shoulder pad styles reduce the linear head accelerations of the struck player. The addition of a foam layer on the shoulder pad further reduces the acceleration. Currently, tethered shoulder pads are the most common style of pads seen on the ice. If hockey equipment manufacturers add a layer of foam padding to tethered pads, the linear head accelerations applied to struck players could drop below half of the concussion level threshold.

Problems & Controversies

In general, controversies within the existing body of research into shoulder-to-head impacts in hockey revolve around the ability of studies to recreate in-game bodychecking and body contact scenarios.

Most biomechanical research into hockey head impacts has been done using one of two methods. The first method is the instrumentation of the helmets of hockey players during gameplay, as was done in the Wilcox et al. study. This method has "strong external validity, but limited accuracy based on current sensor technology". The second method is the attempted recreation of hockey head impacts using laboratory test setups, using instrumented dummies or similar, as was done in the Virani and Richards studies . This method provides high-resolution head acceleration measurements, but due to the extremely unpredictable nature of hockey, it is difficult to recreate player-to-player contact in a realistic manner in a laboratory. As such, most of the existing studies of this type have been conducted in ways that do not accurately represent the types of impacts seen in a real ice hockey game.

For example, in both the Virani et al. and Potvin et al. studies, the body checking dummy was stationary, while in a real game there are scenarios where both the player hitting and the player being hit will have some velocity. With both players moving, the impact scenario would be different, and likely more severe as it may involve rotational or shearing impacts as well as head-on impacts. To determine if design modifications are meaningful in preventing concussions, future studies should account for velocities of both players, as well as oblique or irregular impacts.

There is no consensus among researchers as to which experiment method is more accurate or better at simulating, analyzing, or predicting head injuries in hockey. Some researchers, such as Potvin et al., have developed experiments that combine some aspects of both methods. The Potvin study utilizes a Hybrid III dummy while adding an element of fidelity with bodychecks being delivered by real hockey players, but it still lacks some validity in that the dummy is stationary and upright. Overall, there is still work to be done in this area to make experiments more realistic to better capture the impacts and accelerations that hockey players see during games.

Future Work

Generally, protective sporting equipment is designed to protect the user rather than individuals they come in contact with. Moving forward with scientific studies, as well as the development of sporting equipment, design modifications could be explored to attempt to achieve both of these goals – protecting the user as well as the people who will come in contact with the equipment. With respect to hockey shoulder pads, we suggest that more research be conducted in the field of injury prevention, with a specific focus on materials and construction of the pads.

In the Virani et al. study covering the effect of shoulder pad design on head impact severity, it was found that implementing a 2 cm foam layer to current shoulder pad designs reduced peak linear head acceleration values by 25%, and decreased peak rotational head velocity values by 12.4%. Although this study provided insight on the potential to reduce head impact with changes to shoulder pad design, it could be expanded to provide a clearer guideline for shoulder pad design. This study incorporated a 2 cm foam layer as a single design modification to the pads, but it did not explore or compare any other methods of reducing head impact severity. Future studies should explore the geometry, configuration (size, thickness etc.) of soft and hard layers, and ideal materials to use in shoulder pad design.

Furthermore, the optimization of additional shoulder pad foam thickness and player mobility should be determined. The thicker the foam pad, the more energy absorbed during impact. However, there are thickness limitations when considering player mobility. The thickness of the shoulder pads cannot interfere with the player's ability to stickhandle, pass, and shoot. Additionally, shoulder pad manufacturers need to consider whether this additional foam padding can be retrofitted onto existing shoulder pads or whether a complete redesign is required. Virani et al. only considered the foam padding addition to shoulder pads and did not investigate the effect of adding more foam to helmets or to both helmets and shoulder pads. A potential reason for this is that hockey helmets have already been optimized in size and weight to suit their intended application. Adding more foam to the helmet will create a helmet of larger diameter, which increases the moment applied to the player's neck. Therefore, it could possibly provide a larger risk to the player in some ways.

Although hockey protective equipment has evolved extensively over the years, the current body of research regarding shoulder pads and their effect on head impacts suggests that there is still work to be done to optimize the combination of shoulder pad and helmet modifications to further reduce head impact severity.