Send to

Choose Destination
Am J Sports Med. 2015 May;43(5):1134-41. doi: 10.1177/0363546515570622. Epub 2015 Feb 24.

Association of equipment worn and concussion injury rates in National Collegiate Athletic Association football practices: 2004-2005 to 2008-2009 academic years.

Author information

Datalys Center for Sports Injury Research and Prevention, Indianapolis, Indiana, USA
Datalys Center for Sports Injury Research and Prevention, Indianapolis, Indiana, USA.
University of Arizona, Tucson, Arizona, USA.



The epidemiology of football-related concussions has been extensively examined. However, although football players experience more at-risk exposure time during practices than competitions, there is a dearth of literature examining the nature of the activities or equipment worn during practice. In particular, varying levels of equipment worn during practices may place players at varying levels of risk for concussion.


To describe the epidemiology of NCAA men's football concussions that occurred during practices from the 2004-2005 to 2008-2009 academic years by amount of equipment worn.


Descriptive epidemiology study.


Men's collegiate football data from the National Collegiate Athletic Association Injury Surveillance System (NCAA ISS) during the 5-year study period were analyzed. Injury rates and injury rate ratios (RRs) were reported with 95% confidence intervals.


During the study period, 795 concussions were reported during practices, resulting in an injury rate of 0.39 per 1000 athlete-exposures (AEs) (95% CI, 0.36-0.42). Among NCAA divisions, Division III had the highest concussion rate (0.54/1000 AEs), followed by Division I (0.34/1000 AEs) and Division II (0.24/1000 AEs) (all P values for RRs comparing divisions<.001). Most concussions in practice occurred when players were fully padded (69.9%), followed by wearing shells (23.5%) and helmets only (1.9%). The practice concussion rate was higher in fully padded practices (0.66/1000 AEs) compared with practices when shells were worn (0.33/1000 AEs; RR=1.99 [95% CI, 1.69-2.35]; P<.001) and practices when only helmets were worn (0.03/1000 AEs; RR=22.39 [95% CI, 13.41-37.39]; P<.001). The practice concussion rate of the preseason (0.76/1000 AEs) was higher than that of the regular season (0.18/1000 AEs; RR=4.14 [95% CI, 3.55-4.83]; P<.001) and that of postseason (0.25/1000 AEs; RR=3.02 [95% CI, 1.95-4.67]; P<.001). The types of practice with the highest concussion rate were scrimmages (1.55/1000 AEs). Although only 3 concussions were sustained during scrimmage practices in which players wore shells, the concussion rate (2.84/1000 AEs) was higher than all other reported rates.


Practice concussion rates are highest during fully padded practices, preseason practices, and scrimmages, suggesting that the nature, focus, and intensity of football practices affect concussion risk. In addition, coaching staff should continue to closely monitor player safety during scrimmages. Meanwhile, future surveillance should examine whether removing scrimmages, particularly those that are not fully padded, will meaningfully reduce the incidence and rate of concussions.


epidemiology; injury prevention; scrimmages; traumatic brain injury

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center