The Physician and Sportsmedicine
Menubar Home Journal Personal Health Resource Center CME Advertiser Services About Us


Hitting in Amateur Ice Hockey: Not Worth the Risk

William O. Roberts, MD


Like it or not, hitting is the soul of hockey, and there's nothing like a resounding body check to change the flow of a game or even the course of history.—writer Michael Farber, in a 1999 Sports Illustrated article titled "Check, Please" (1)

I can hit anyone, anytime, as hard as I want, and it doesn't hurt me.—A 15-year-old Bantam level hockey player

The "big hit" is glorified in professional ice hockey. But body checking is unnecessary for the quality of the game and demonstrably dangerous for most amateur players at the youth, high school, and college levels.

Ice hockey is a collision sport, but the intensity of body contact delivered and permitted has escalated beyond the margin of safety. Violent body collisions in professional ice hockey relate more to the entertainment value of the event than the true beauty of the sport, but, unfortunately, the style of play at all age levels is influenced by the professional leagues. Too often, the major swings in the flow of the game caused by the "big hit" may be due to player injury, especially at the youth level.

The profile of hockey injuries has changed over the years. In the 1970s, facial lacerations and dental injuries predominated, and cervical-spine injuries were rare. Now, following equipment changes adding helmets, full face shields, and heavy padding, facial and dental injuries are almost nonexistent, but catastrophic cervical-spine injury is a fixture of the hockey scene, and a host of other injuries—including shoulder dislocation, fractures, and concussion—are common. A review of the literature demonstrates these trends and suggests direction for the future.

Youth Injury Rates

Community-based studies. Several colleagues and I began to look at youth ice hockey injury during the 1989-90 season (2). We tracked 152 boys in nine youth teams (ages 9 to 15), at three age levels (Squirt, Peewee, and Bantam) and three ability levels (A, most skilled; B, intermediate skilled; and C, least skilled) in a Minnesota community to assess the rate, severity, and distribution of injuries. We defined injury as any interruption in play. The injury rate was 15 per 100 players per season (52 total injuries, including 8 fractures). Thirty-eight players were injured during games as a result of body contact; 9 injuries involved legal checking, and the remainder resulted from illegal checks or other violations. Eighty-eight percent of injuries involved collisions.

Stuart et al (3) studied the same age-groups in a different Minnesota community at the A level of play during the 1993-94 season. The injury definition was restricted to injuries lasting more than 1 day or involving concussion, skin laceration, or dental injury (National Athletic Injuries/Illness Reporting Systems [NAIRS] definition). The study of 66 players showed 21 injuries per 100 players per season (14 injuries reported), with game injury rates of 0, 0, and 10.9 injuries per 1,000 player-hours for Squirt, Peewee, and Bantam teams, respectively. The majority of injuries in this study were also game related. Four of the injuries were fractures; 50% involved collisions.

We reanalyzed the data from our study (2) using the NAIRS injury definition and the same exposure assumptions as used by Stuart et al. The game injury rates for the Squirt A, Peewee A, and Bantam A boys' teams were 6.7, 12.3, and 10.8 injuries per 1,000 player-hours.

In a 1993 Finnish study of 1,437 players aged 9 to 18, Bjorkenheim et al (4) reported 9 injuries per 100 players per season, 9% of which sidelined athletes longer than 1 day. Ninety-four percent of the injuries occurred during games, and 53% were due to collisions.

To build on our data from the community-based season studies, my colleagues and I looked at game injuries during youth tournaments involving 807 boys in the 1993-94 season (5). Using on-site athletic trainers to record injuries, we found that the game injury rate was four to five times greater during tournament play than during regular-season play at the Bantam and Peewee levels. For example, during Bantam A competition, the tournament injury rate was 43 per 1,000 player-hours, compared with 11 per 1,000 player-hours in the two community studies (2,3). In tournaments, collision caused 65% (39 of 60) of total injuries and 77% (20 of 26) of NAIRS-definition injuries (5). Eleven penalties were associated with injuries; 6 were for checking from behind. As in our initial study (2), the majority of players returned to the same game or returned by the next day. There were no injuries in girls' competition, where body checking is banned.

In one survey (6), associates and I compared injury rates during the regular-rules and "fair-play" portions of a junior tournament. The injury rate was five times greater and the number of penalties assessed was two times greater in the regular games than in the fair-play games.

College and elite amateur injuries. Injuries in male college hockey are similar in frequency and pattern to youth hockey. A recent 3-year study (7) of nine National Collegiate Athletic Association Division 1 teams reported a game injury rate of 14.7 per 1,000 athlete-exposures using criteria similar to NAIRS reporting. Sixty-five percent of the injuries involved collision, and there were 18 fractures with a mean time lost of 22.2 days. Collision with other players, the ice, and the boards accounted for the greatest number of days lost, and most of the injured players missed at least 1 week of competition or practice.

A survey (8) of injuries at the 1997 Ice Hockey World Championships defined injury as any sudden trauma during games requiring treatment by a physician or dentist. The total injury rate was 166 per 1,000 player-hours. The rules did not require face protection, and when the 30 facial injuries are factored out, the adjusted injury rate was 113 injuries per 1,000 player-hours. Thirty-one players were unable to continue play, giving an adjusted injury rate of 48.7 per 1,000 player-hours. The fracture rate was 9.4 per 1,000 player-hours. Through the use of videotaping, this study also demonstrated that games in which officials enforced the rules more rigidly had lower injury rates.

Discussion. It is apparent that tournament and higher-level play is associated with higher injury rates. It is also clear that in both tournament and regular-season play, collision accounts for most injuries in all of the youth hockey studies. Most of these result from body checking, which is more vigorous during competition than in practice. Eliminating intentional high-energy body collision would make the youth game safer.

There is also evidence that better compliance with game rules prevents injuries. Our research (and that of Tator et al, below) has found that a high percentage of injuries are caused by rule violations (5). We also observe that the fair-play system, in which teams are rewarded for good behavior with points in the standings, clearly reduces injury rates (6).

Neurologic Concerns

Concussion. Cerebral concussion or mild brain injury rates vary among the studies. In season-length youth hockey studies, an early study by Sutherland (9) that included players as young as 5 recorded 0.09 concussions per 1,000 player-hours, we (2) estimated 0.75 per 1,000 player-hours, and Stuart et al (3) reported no concussions. A high school hockey study (10) reported a rate of 10 per 100 players per season. In our youth tournament study (5), concussion rates ranged from 10.7 to 23.1 per 1,000 player-hours. As with other injuries, most concussions are caused by collision.

Catastrophic injury. Prior to the use of modern protective equipment, cervical-spine injury was rare in hockey. The work of Tator et al (11,12) documents the change in Canada. Tator's registry (11) now has 215 cases of neck fracture and has leveled off at 15 cases per year; the majority of these injuries are caused by an illegal check from behind (12).

In the United States, the data of Cantu and Mueller (13) demonstrate a direct fatality and injury rate of 3.11 per 100,000 high school hockey players and 11.55 per 100,000 at the college level. Most of the injuries occur when the athlete is struck from behind or falls and hits the boards. Data for USA Hockey participants are not published.

Discussion. Because the current helmet does not completely protect the player from minimal brain trauma, and because data on the cognitive effects of repeated concussion continue to accumulate (14), concussion rates are of increasing concern. Like other injuries, concussion is caused primarily by collision, and concussion rates appear to increase with competition level; the Stuart et al study (3) may have missed minor concussions, which by definition resolve quickly. Concussion rates during tournaments, in particular, are very high compared with rates for season play.

The need for attention to the issue of serious injury in hockey is made evident by a comparison. In football, widely perceived to be hazardous, the direct fatality and injury rates are half those in hockey: 1.76 per 100,000 football players in high school and 6.96 per 100,000 in college (13).

From the data on mechanism of injury, particularly for cervical-spine injury, it's also clear that it's essential to strictly enforce the check-from-behind rule and to continue education of players and coaches about the risk of cervical-spine fracture. Maximal decreases in injury rates are more likely with a two-pronged approach: eliminate checking and strictly enforce the rules.

Aggression, Infractions, and Injuries

Players today are more skilled, bigger, stronger, and faster than they were a generation ago, and they are wearing padding and other equipment that decreases the risk that an aggressive maneuver will cause pain. (My 15-year-old patient, quoted at the beginning of this article, states the case succinctly.) As a result, players collide with greater force.

Furthermore, a vicious circle results. Body checking leads to a relaxed attitude toward collision and thus contributes to an increase in rule infractions (15). Since the players tolerate more aggression and more intense collisions, the officials are forced to control the game more loosely to keep 10 skaters on the ice.

Although rule infractions don't necessarily follow from body checking, it's my observation that one encourages the other.

Time to Check Checking

The simplest solution, then, to the high rate of injury in ice hockey is to follow the example of the women's and girl's game and eliminate body checking, at least for the youth levels of the sport. I'm not alone in recommending this solution: For example, in 1991, the Canadian Academy of Sport Medicine (16) recommended that intentional body contact not be allowed except in the preinternational and preprofessional leagues.

Would ice hockey suffer without checking? The women's game is remarkably free of injury and is still exciting to watch. In fact, many players of a generation ago bemoan the current emphasis on the body check, remembering their own reliance on speed and puck-handling skill. Eliminating checking would allow young players to concentrate on these aspects of the game. The very small percentage of ice hockey players who move to the professional level of play could certainly learn the art of body checking in short order.

Helmets, full face shields, and maximal padding reduce injuries in ice hockey (17,18), but safety equipment can't yield maximum protection without effective enforcement of the rules. Furthermore, it is unlikely that male hockey can return to stricter rule enforcement without a strong effort by players, coaches, and referees to tone down the intensity of the body collisions. A ban on body checking, backed by fair-play rules, would be a major step in that direction.


  1. Farber M: Check, please. Sports Illus 1999;90(13):46-51
  2. Brust JD, Leonard BJ, Pheley A, et al: Children's ice hockey injuries. Am J Dis Child 1992;146(6):741-747
  3. Stuart MJ, Smith AM, Nieva JJ, et al: Injuries in youth ice hockey: a pilot surveillance strategy. Mayo Clin Proc 1995;70(4):350-356
  4. Bjorkenheim JM, Syvahuoko I, Rosenberg PH: Injuries in competitive junior ice-hockey: 1437 players followed for one season. Acta Orthop Scand 1993;64(4):459-461
  5. Roberts WO, Brust JD, Leonard B: Youth ice hockey tournament injuries: rates and patterns compared to season play. Med Sci Sports Exerc 1999;31(1):46-51
  6. Roberts WO, Brust JD, Leonard B, et al: Fair play rules and injury reduction in ice hockey. Arch Pediatr Adolesc Med 1996;150(2):140-145
  7. Ferrara MS, Schurr KT: Intercollegiate ice hockey injuries: a causal analysis. Clin J Sport Med 1999;9(1):30-33
  8. Mölsä JJ, Tuominen M, Väisänen J, et al: Injuries in ice hockey championships 1997. Med Sci Sports Exerc 1999;30(5 suppl):S400
  9. Sutherland G: Fire on Ice. Am J Sports Med 1976;4(6):264-269
  10. Gerberich SG, Finke R, Madden M, et al: An epidemiological study of high school ice hockey injuries. Childs Nerv Syst 1987;3(2):59-64
  11. Tator CH, Carson JD, Edmonds VE: New spinal injuries in hockey. Clin J Sport Med 1997;7(1):17-21
  12. Tator CH, Edmonds VE, Lapczak L, et al: Spinal injuries in ice hockey players, 1966-1987. Can J Surg 1991;34(1):63-69
  13. Cantu RC, Mueller FO: Fatalities and catastrophic injuries in high school and college sports, 1982-1997: lessons for improving safety. Phys Sportsmed 1999;27(8):35-48
  14. Collins MW, Grindel SH, Lovell MR, et al: Relationship between concussion and neuropsychological performance in college football players. JAMA 1999;282(10):964-970
  15. Parayre R: The effects of rules and officiating on the occurrence and prevention of injuries, in Castaldi CR, Hoemer EF (eds): Safety in Ice Hockey, ASTM STP 1050. Philadelphia, American Society for Testing and Materials, 1989, pp 37-43
  16. Canadian Academy of Sport Medicine: Position statement: violence and injuries in ice hockey. Clin J Sport Med 1991;1:141-144
  17. Benson BW, Meeuwisse WH, Mohtadi NGH: Head and neck injury risk in hockey players wearing full versus half shields: preliminary data, abstracted. Med Sci Sports Exerc 1998;30(5 suppl):S157
  18. Benson BW, Meeuwisse WH, Mohtadi NGH, et al: Head and neck injury risk among intercollegiate ice hockey players wearing full versus half face shields, abstracted. Clin J Sport Med 1999;9(2):108

Dr Roberts is a former amateur hockey player, a family physician at MinnHealth SportsCare in White Bear Lake, Minnesota, and medical director of the Twin Cities Marathon. He is a fellow of the American College of Sports Medicine and a senior associate editorial board member of The Physician and Sportsmedicine. Address correspondence to William O. Roberts, MD, MinnHealth SportsCare, 4786 Banning Ave, White Bear Lake, MN 55110.


The McGraw-Hill Companies Gradient

Copyright (C) 1999. The McGraw-Hill Companies. All Rights Reserved
Privacy Policy.   Privacy Notice.