Background There is a paucity of literature examining injury and illness rates in men's professional ice hockey. This study aimed to determine injury and illness rates in the NHL over six seasons, and identify predictors of injury-related time loss in this population.

Methods This study involved an inclusive cohort of hockey players from all NHL teams competing in the 2006–2007 through 2011–2012 seasons. A standardised electronic injury surveillance system was used to report injury and illness events. The primary outcome was regular season and postseason time-loss injury/illness. The secondary outcome was man-games lost from the competition.

Results On the basis of the estimated athlete exposures (AEs), the overall regular season incidence density was 15.6 injuries/1000 AEs and 0.7 illnesses/1000 AEs. Based on recorded time on ice, the injury rates were roughly threefold higher at 49.4 injuries/1000 player game-hours and 2.4 illnesses/1000 player game-hours. There was a reduction in injury rates over the 6-year period, with the greatest reduction between the 2007–2008 and 2008–2009 seasons. Multivariate predictors of time loss greater than 10 days were being a goalie (OR=1.68, 95% CI 1.18 to 2.38), being injured in a road game (OR=1.43, 95% CI 1.25 to 1.63) and the mechanism of injury being a body check (OR=2.21, 95% CI 1.86 to 2.62).

Conclusions There was an overall reduction in the time-loss injury and illness rates over six seasons. Being a goaltender, being injured on the road and being injured by a body check were the risk factors for time loss greater than five ‘man games’.