Background In 2013, Hockey Canada announced a national policy change delaying the introduction of body checking in ice hockey until Bantam (ages 13–14). Following this change, several local organizations in Canada changed policy to also disallow body checking in Bantam non-elite levels (lower 60–70% by division of play).

Objective To determine if the risk of concussion and severe concussion differs for non-elite Bantam ice hockey players in leagues where body checking is permitted compared with leagues where local policy change removed body checking.

Design Cohort study.

Setting Ice hockey rinks (2014/15, 2015/16 seasons).

Participants Youth ice hockey players in non-elite Bantam body checking leagues (52 teams, 653 players) in Calgary and Edmonton (2014/15) and non-body checking leagues (33 teams, 396 players) in Kelowna and Vancouver (2014/15) and Calgary (2015/16), Canada.

Assessment of Risk Factors Exposure to a Bantam league where body checking was not allowed.

Main Outcome Measurements Suspected concussions were identified by a team designate and referred to a sport medicine physician. Severe concussions were defined by greater than 10 days of time-loss from ice hockey.

Results There were 47 game-related concussions (Incidence Proportion per 100 players [IP] =7.2; 95% CI; 5.33-9.46) and 27 severe concussions (IP=4.1; 95% CI; 2.74-5.96) in body checking leagues. In non-body checking leagues there were 12 concussions (IP=3.03; 95% CI; 1.58-5.23) and 8 severe concussions (IP=2.02; 95% CI; 0.87-3.94). There was no evidence of a difference between the non-body checking leagues in Kelowna and Vancouver (2014/15) and Calgary (2015/16). Preliminary univariate analysis indicates a protective effect of policy disallowing body checking [risk ratio (RR)=0.42 (95% CI; 0.2-0.88) for concussion and RR=0.49 (95% CI; 0.23-1.04) for severe concussion.

Conclusions This preliminary analysis suggests a 58% lower risk of game-related concussion in non-elite Bantam leagues where local policy disallows body checking.