Study shows where NHL can improve concussion prevention

Saturday July 20th, 2013

Sidney Crosby, the NHL's marquee star, became the poster boy for its concussion epidemic.
Damian Strohmeyer/SI

The publication last week of a scholarly report on concussions in the NHL and OHL made headlines for its conclusion: Stronger rules on hits to the head have done nothing to decrease concussions.

It's a bold assertion and the way the data was organized and presented leaves little room for doubt. But let's stroll into that little room and see what's there, because I suspect there's something more that needs to be said.

First, when you look at the report -- "Bodychecking Rules and Concussions in Elite Hockey" by Laura Donaldson, Mark Asbridge and Michael D. Cusimano on the website of the open access scientific journal PLOS one -- get set to be somewhat perplexed by the authors' methodology. Unless you have a deep background in statistical analysis, you'll likely zone out when you see sentences like, "One- and two-way analysis of variance (ANOVA) with post-hoc comparisons using Bonferroni corrections were used to determine changes over the three seasons analyzed," and terms like "Chi-squared" and "Fisher's exact tests."

You're going to say to yourself, "What the Bonferroni does that have to do with Sidney Crosby missing a quarter of the season?"

Now, I've had a fair bit of education, but none in statistics. I have zero idea what any of that means and couldn't tell you even if I had stayed at a Holiday Inn Express last night. So even if the authors' math techniques are flawed (I'm going to assume they're not) and the sample size they used is not too small, they've done a masterful job assembling their numbers and constructing some fine charts to display their data and give it meaning.

Here's what they came up with: The NHL and the OHL both have some relatively new rules that make certain types of checks to the head illegal. But after examining a sample of games from both leagues over the last three seasons -- the same 10 weeks in both leagues -- the authors found no appreciable difference in the number of concussions, confirmed or suspected, after the rules went into effect. Their analysis showed that for every 100 NHL games played, 5.23 players suffered concussions (in the OHL, it was 5.05 per 100) and during the course of the study, there was little change in that rate from year-to-year.

For the NHL, this means there were pretty much the same number of concussions in 2009-10 as there were the next season when the first version of Rule 48 outlawed blindside and lateral hits to the head. And even during the year after that, 2011-12, when Rule 48 was strengthened to include any hit in which the head is targeted or the principle point of contact, there were the same number of concussions.

Well, that sounds like nothing but bad news, and when Cusimano, a very highly regarded neurosurgeon who leads the injury prevention unit of St. Michael's Hospital in Toronto, was interviewed by Canadian Press, he concluded, "If player safety is the prime priority of the NHL in bringing this kind of rule in ... then they need to relook at this in a very serious way and adjust things. If it isn't a priority, I could see them just leaving things the way they are and it's kind of a Band-Aid response to a major problem."

Tough to argue with that sentiment.

But let's back up for a second: Rule 48, in its original form, was not intended to eliminate all concussions from hockey only blindside hits to the head. It was limited, something the league recognized by improving the rule a year later. It was, however, a start. And it has largely worked. Even the study's own stats demonstrate that. The authors cited a drop in concussions caused by blindside hits in 2011-12 to none, but didn't notice that many to begin with: only three of the 26 concussions they tracked in their sampling of NHL action in the year prior to Rule 48's existence, and only two among the 55 that were noted in their sampling from the first year of Rule 48. In fact, the study dismisses the blindside hit as a major cause of concussions, attributing only 4.1 percent of total cases to such instances. "NHL concussions did not decrease after the introduction of Rule 48 in 2010-11, which may be unsurprising given that blindsiding was found to be an uncommon cause of NHL concussions," they write.

But perhaps they shouldn't be so dismissive because the saga of Rule 48 actually makes the authors' argument for stronger rules more persuasive. As an NHL spokesman told, the league's data for the entire 2012-13 schedule shows that concussions from blindside and lateral hits have pretty much vanished from the game. During a conversation that I had with NHL vice president Brendan Shanahan last spring, he mentioned that he now routinely sees NHL players pull away from making that hit when they see a vulnerable puck carrier. So Rule 48 has worked. It has done what it was intended to do: get that hit out of the game.

There was such great uproar -- and there still is -- about blindside hits that we tend to lose focus on how the vast majority of concussions are caused. It's something the league has pointed out and this study reinforces the claim that most concussions do not result from illegal hits. "Illegal incidents, where the aggressor was assessed a penalty, fine or suspension, accounted for 28.4 percent of cases for concussion and 36.8 percent of cases for suspected concussions," the study found. So, most of the damage (and suspected damage) is done by hits that aren't against the rules.

That, in part, speaks to the fact that some rules may need to be tougher. It also speaks to the nature of the game.

While the majority of concussions, more than 64 percent, came from bodychecks, many were from hits that didn't even involve head contact. Over the course of its three years, the study found 39 concussions resulting from checks in which at least some contact was made to the head. Meanwhile, 35 were suffered when there was no head contact, and another five were caused by checks during which head contact could not be conclusively determined.

And that's not all. "Unintentional actions such as a collision with a teammate were responsible for 4.9 percent of cases and an additional 12.2 percent were caused when players were hit by the puck," the authors found. Additionally, "51.2 percent of all incidents involved a secondary contact of the head after the initial impact, most commonly to the boards or ice. Being hit in the face by a puck was the most common cause of suspected concussion associated with facial fracture (7 of 12 cases). Most of the players injured in this manner were not wearing a visor at the time (6 of 7 cases)."

A player's reputation is a factor in punishing dangerous hits, but long suspensions remain rare.
Chris Williams/Icon SMI

All that just reinforces the belief this is a rough sport, a collision sport, with hard things like ice, boards, sticks and pucks and other players all being potential concussion-causing factors. Without trying to sound terribly facetious, short of padding the boards, making the ice into pudding and using foam rubber sticks and pucks, the playing environment will always have inherent dangers. That's just the nature of hockey.

And here we can credit the NHL for mandating rounded and padded glass supports in all rinks in recent years and -- finally, after at least 15 years of doing very little -- eliminating the unforgiving seamless glass systems in all its arenas.

Another long called-for measure, making visors mandatory, goes into effect this coming season, albeit with a grandfather clause, and from the study's data, that's bound to be another small improvement.

That leaves one area that will continue to require concern, study and action: other players. First, the protective equipment they wear needs to be addressed. Hard elbow and shoulder pads have frequently been blamed as potential causes of concussion, but the league and the NHLPA can't seem to find a way to soften them.

But most of the attention gets focused on the rules of bodychecking, as perhaps it should since checking is the biggest cause of concussions. Again, when considering the high number of concussions caused by checks that don't involve head contact, we're talking about the nature of the game. As the study notes, there's the secondary contact factor as well as things like whiplash when a hard check is delivered in such a way that it causes the puck carrier's head to move forcefully even without direct contact, and his brain still makes contact with his skull and is traumatized. The only way to eliminate that is to remove bodychecking as a whole from the NHL, making hockey a non-contact sport. That's as unlikely as the NFL doing away with blocking and tackling.

Hockey is becoming more and more a game of hits. NHL figures bear that out: In 2005-06, the league counted 29,305 hits during the regular season, an average of 32 per game. That number has trended continually upward. In 2011-12, the last full 82-game season and (the third year of the study), there were 55,445 hits, an average of 45 per game, both new high water marks. The average of 47 per game during the lockout-shortened 2013 schedule projects to 57,810 over a full 1,215-game schedule, and that would have been an all-time high. This is what the NHL is today. Players are going to be hit. Some of those hits will hurt, and some will result in concussions. There's no avoiding that.

Which is not to say that the league shouldn't do everything possible to create a safer game within that physical context. From there, we have to separate out the bodychecks in which the head is targeted. While the improved Rule 48 does that to an extent, it still allows the referees some leeway if the puck carrier puts himself in a vulnerable position.

"So it's like his fault, because he put himself into a vulnerable position," Cusimano told Canadian Press. "And this highlights one of the major problems in sport and particularly in hockey these days. We victimize the victim even more, rather than looking at the game and the system and saying: 'What can we do to reduce these injuries?"'

That loophole is closed in amateur hockey throughout the world and in European pro hockey as well. There's certainly resistance to implementing the tougher rule in the North America pro game, but it's worthy of re-examination.

There's also the issue of enforcement and how lenient both on-ice officials and the Department of Player Safety can be in penalizing perpetrators. It seems as if a player needs to be a multiple offender, like Raffi Torres, before the league really slams him with a meaningful suspension. Many observers have noted that there is little sentiment in the hockey community for harsher player discipline. As a whole, GMs, coaches, owners, players, the union and agents oppose longer suspensions and not much is going to change unless their position does.

And it's well known that an injury or the lack of one will play a major role in determining if a player is punished for delivering a dangerous hit. There continues to be a sense that the league looks for reasons to excuse players more readily than it tries to set a tone that would keep them more mindful of the consequences of such play.

A perfect example is Rick Nash's hit on Tomas Kopecky last March.

No penalty was called on the play, and while Shanahan ruled that one should have been, he declined to suspend Nash, saying the head was not the principle point of contact. Here's his decision:

Perhaps it wasn't a Rule 48 violation, but it certainly was charging and it propelled Kopecky's helmet off his head as he fell to the ice. It's a bad hit, even though it might have been unintentional. And, as the study mentions, there are many hits that are considered legal that still result in concussions, and secondary contact can cause injury. Shouldn't the NHL do more to try limiting these instances? This one play would have been considered legal in the study, but really wasn't -- and it could have had a worse outcome, but it didn't. A tougher standard is really called for in these situations.

"Part of it's the way the rule's written," Cusimano said. "Part of it's the way the rule is enforced. Part of it's the penalties associated with the rule. And part of it is that concussions are also coming from other causes like fighting, that is still allowed."

It is inevitable that fighting has to be part of this discussion, but the study found it was not a major contributor to concussions. Of the 123 noted during the three-year course of the study, 11 were attributed to fights. For some people, that's 11 too many.

The bottom line is the same as it's always been: Some things about the game can't be changed without changing the game itself. Figuring out what those things are can itself be a contentious matter -- remember those handwringers who swore the first version of Rule 48 would ruin the sport? It didn't happen.

But if the NHL truly wants certain sorts of behavior out of the game, it can be accomplished. Things rarely change overnight in any case, but they don't change at all if people don't search for solutions. If the hockey community wants to reduce concussions, this study at least shows the areas in which it should be searching.

SI Apps
We've Got Apps Too
Get expert analysis, unrivaled access, and the award-winning storytelling only SI can provide - from Peter King, Tom Verducci, Lee Jenkins, Seth Davis, and more - delivered straight to you, along with up-to-the-minute news and live scores.