OBJECTIVE: National Football League game video was analyzed for the typical locations of severe helmet impacts in professional football. By use of selected cases that were reconstructed in laboratory tests and reported previously, the magnitude and direction of force causing concussion was determined for these locations.
METHODS: Multiple video views were obtained for 182 severe helmet impacts that occurred between 1996 and 2001. From a top view, the helmet was divided into 45-degree quadrants with 0 degrees eyes forward. From a side view, it was divided into seven equal levels, four (+Q1 to +Q4) above the head center of gravity and three below (−Q1 to −Q3). The initial helmet contact was located in these regions. Thirty-one impacts were reconstructed with helmeted Hybrid III dummies involving 25 concussions. Measurement of head translational and rotational acceleration was used to determine the average and ±1 standard deviation in responses, with impacts reflected to the right side.
RESULTS: From video, the majority (71%) of impact is to the helmet shell primarily from a striking player’s helmet, arm, or shoulder pad to the side (45–135 degrees) or from ground contact to the back (135–180 degrees). Most impacts were high on the helmet at +Q2 to +Q4. The remainder (29%) were primarily from helmet contact on the facemask at an oblique frontal angle (0–45 degrees) and −Q3 to +Q1 height. From reconstructions, concussion occurred with the lowest peak head acceleration in facemask impacts at 78 ± 18 g versus an average 107 to 117 g for impacts on other quadrants (t = 2.90, P < 0.005). There was a significantly higher head acceleration for concussed versus nonconcussed players (t = 2.85, P < 0.05). The vector of peak force was essentially horizontal for facemask impacts and downward at 12 to 27 degrees for impacts to the helmet side and back. Concussion in professional football involves four typical conditions, as follows: A, 0- to 45-degree quadrant, −Q3 to +Q3 level, peak force 49 ± 18 degrees from front and horizontal; B, 45- to 90-degree quadrant, −Q2 to +Q3 level, peak force 73 ± 12 degrees and horizontal; C, 90- to 135-degree quadrant, +Q1 to +Q4 level, peak force 97 ± 9 degrees and 12 degrees downward; and D, 135- to 180-degree quadrant, +Q1 to +Q4 level, peak force 157 ± 1 degrees and 27 degrees downward. Concussed players averaged 3.6 ± 2.7 initial signs and symptoms. The most common were headaches, dizziness, immediate recall problems, and difficulty with information processing.
CONCLUSION: The location, direction, and severity of helmet impacts causing concussion in the National Football League have been defined from analysis of game video and laboratory reconstruction. These conditions define the circumstances in which helmets need to reduce head injury risks in professional football.
In 1994, the Commissioner of the National Football League (NFL) formed a Committee on Mild Traumatic Brain Injuries (MTBI) in response to safety concerns regarding concussion in professional football (9). After a number of years of multifaceted research on concussion biomechanics, injury epidemiology, neuropsychological testing, brain response modeling, and treatment, the Committee has started publishing its scientific findings.
The first published study involved the initial findings on concussion biomechanics in professional football (10). It analyzed game video of concussion and severe head impacts and selectively reconstructed impacts with Hybrid III dummies to measure head accelerations with concussion. The work showed a considerable severity of concussion impacts in NFL football, with an average 9.3-m/s (20.8-mph) impact velocity, 7.2-m/s (16.1-mph) change in head velocity (head ΔV), 98 g head acceleration, and 15-ms duration. The peak force of concussion averaged 4.4 ± 1.2 kN (980 ± 280 lb) on the head.
The findings of the first study are extended by further analysis of the laboratory reconstruction data and the video database of severe helmet impacts in NFL games from 1996 to 2001. The game video showed impact locations on the helmet, which were classified by dividing the helmet into four quadrants from a top view. Data from left-side impacts were reflected to the right side of the helmet for analysis. This used 45-degree quadrant angles from the front (0 degrees eyes forward) to the rear (180 degrees). From the side, the helmet was divided into four equal levels above and three below the head center of gravity (CG).
The purpose of classifying the initial helmet contacts was to determine whether impact clusters could be defined for the most typical conditions causing concussion. This would allow the determination of average responses for the most important clusters and refine the understanding of typical conditions causing concussion in NFL players. By adding biomechanical response data, the picture of helmet impacts would be further clarified.
The aim of this work was to define conditions representing the majority of concussion in professional football so that current helmet performance can be assessed and efforts taken to reduce risks of concussion in game impacts. This would motivate improvements in helmet performance for the critical impact conditions, help the National Operating Committee on Standards for Athletic Equipment (NOCSAE) establish new standards to evaluate helmets, and improve the safety of professional football players and others.
Mild Traumatic Brain Injury Committee, National Football League, New York, New York, and ProHEALTH Care Associates, LLP, Lake Success, New York
Mild Traumatic Brain Injury Committee, National Football League, New York, New York, and ProBiomechanics LLC, Bloomfield Hills, Michigan
Mild Traumatic Brain Injury Committee, National Football League, New York, New York, and University of Maryland School of Medicine, Timonium, Maryland
Mild Traumatic Brain Injury Committee, National Football League, New York, New York; Department of Neurology, Long Island Jewish Medical Center, New Hyde Park, New York; and Department of Neurology, New York University Medical Center, New York, New York
David C. Viano, Dr. med., Ph.D., ProBiomechanics LLC, 265 Warrington Road, Bloomfield Hills, MI 48304-2952. Email: firstname.lastname@example.org
Received, June 10, 2003.
Accepted, August 14, 2003.