That all important denominator for injury rates: time or exposure based?

By Jon Candy, via Flickr

Doing an injury surveillance study may begin with simply counting the number of injuries (that, ideally, meets the most widely adopted definition of injury set forth by Fuller, et al., in Brit J Sports Med 2006; 40:193-201). But that just gives the frequency of injury during some defined time period.

The more important figure is the rate of injury. While there are a number of different ways to report rate, the two most common methods are the number of injuries per 1000 player-hours or per 1000 athlete-exposures; the first is based on the amount of time a player is exposed to an injury risk and the later is just the number of exposures to an injury risk. The denominator is very important.

Stovitz and Shirer argue that the time exposed to risk for game (or training) injuries in sports is best assessed with actual minutes of play per athlete. For example, if a central defender plays 60 minutes and is then substituted out, the player was exposed to 1 hour of play and the sub was exposed to .5 hour of play but each had one athlete-exposure, which does not consider the duration of exposure. Individual-level playing time, however, is often unavailable so sports injury epidemiologists will use one of two basic methods for the amount of time an athlete is exposed to risk during training or during competition. Some will count only the number of athletes on the field during the game of the particular sport.

In the example above, the central defender (the position, not a particular player) was exposed to 1.5 hours of play even though there were 2 different people playing the position. This method approximates individual exposure time if the project had full team enrollment and games are played with a consistent number of players; the same 11 who started the match also finished the match. Thus, 11 players exposed to a 1.5-hour match accumulated 16.5 match-hours of exposure (11 x 1.5). This 16.5-hour sum for a 90min, 11v11 football match is a constant (ignoring added extra time or overtime, but would need to be accommodated). By individual player, however, the starter had 1 match-hour of exposure and the sub had .5 match hours of exposure and this would probably be the most accurate component of a denominator, but lacking actual playing time makes this a rare method indeed.

Other researchers apply one unit of exposure time to every member of a team who plays in a game regardless of the number of minutes played. Thus, 11 players start and finish the match accumulate 11 match-exposures. If a team has 3 subs, all in the last 15 minutes of the match accumulates 14 match-exposures. Some even give a match exposure to every player on the match day roster, even those who don’t play, so if there were 7 subs, the team would have 18 match-exposures. This latter method underestimates game injury rates by over overestimating exposures. These are important distinctions when reading