Determination of athlete-exposures in team sports: it depends on the research question

Abstract

We would like to introduce you to Sarah, a sports medicine clinician who cares for team sports and is interested in decreasing injury rates. Sarah recognises that to decrease injury rates, she must properly define the term and then use the best methods for assessment. Epidemiology includes definitions for measures such as incidence rates,1 and Knowles et al applied these principles to sports injuries in 2006.2 Injury rates depend on a proper numerator, that is, the injuries, and a proper denominator, that is, exposure time. Our focus is on the denominator, exposure time, where the assessment should only include ‘athletes-at-risk’.2 In sports, an athlete is generally only at risk for injury while they are on the field, playing their sport. This editorial focuses on team sports where Sarah notices that many injury surveillance studies calculate exposure time in team sports using the phrase, ‘athlete-exposures’, where it is defined as ‘one athlete participating in one game or practice’.3–6 This definition is imprecise and some have interpreted ‘participating’ as only those who played in the games, regardless of the duration,7 and others have interpreted it to include everyone on the game roster, even if they had no playing time.8 Since players on the bench are not at risk for a game-time injury, if there are any substitutions, then both of the above definitions of ‘athlete-exposures’ violate the rule to assess only those at risk. Therefore, for many research questions, these definitions of athlete-exposures overestimate exposure time and consequently underestimate injury rates. This is especially important when comparing across sports as some sports have a large percentage of players on the bench as opposed to in the game, for example, basketball and ice hockey, whereas other sports may have few players on the bench compared with in …