Using video analysis for concussion surveillance in Australian football

Abstract

Objectives

The objectives of the study were to assess the relationship between various player and game factors and risk of concussion; and to assess the reliability of video analysis for mechanistic assessment of concussion in Australian football.

Design

Prospective cohort study.

Methods

All impacts and collisions resulting in concussion were identified during the 2011 Australian Football League season. An extensive list of factors for assessment was created based upon previous analysis of concussion in Australian Football League and expert opinions. The authors independently reviewed the video clips and correlation for each factor was examined.

Results

A total of 82 concussions were reported in 194 games (rate: 8.7 concussions per 1000 match hours; 95% confidence interval: 6.9–10.5). Player demographics and game variables such as venue, timing of the game (day, night or twilight), quarter, travel status (home or interstate) or score margin did not demonstrate a significant relationship with risk of concussion; although a higher percentage of concussions occurred in the first 5 min of game time of the quarter (36.6%), when compared to the last 5 min (20.7%). Variables with good inter-rater agreement included position on the ground, circumstances of the injury and cause of the impact. The remainder of the variables assessed had fair-poor inter-rater agreement. Common problems included insufficient or poor quality video and interpretation issues related to the definitions used.

Conclusions

Clear definitions and good quality video from multiple camera angles are required to improve the utility of video analysis for concussion surveillance in Australian football.

Introduction

Sports concussion is an important issue in active communities. In the United States, data from the Centers for Disease Control and Prevention estimate that 1.6–3.8 million concussions occur annually from participation in sport and recreational activities.¹ In Australia, rates of hospitalisation for sports concussion have increased by 60.5% over a 9 year period from 2002.² From a clinical perspective, recognising the injury and removing the player from training or competition is thought to be important in reducing the risk of adverse outcomes following concussion.³ While professional sports have moved to enhance their concussion recognition and management programs, knowledge translation to the majority of participants who compete in sub-elite, community and junior competitions, remains an ongoing challenge.4, 5, 6

In addition to a conservative approach, a comprehensive concussion management program must include measures to reduce the risk of head injury. Common preventative measures, such as the use of helmets or protective equipment, have limited evidence for their efficacy⁷ and are not routinely used in all sports. However, understanding the circumstances and mechanism of injury unique to each sport may facilitate development of strategies to reduce concussion risk.⁸ For example, stricter interpretation of the rules was demonstrated to be effective in reducing the number of head injuries related to raised elbows in heading contests in Norwegian football.9, 10

Video analysis has successfully been used to investigate the biomechanics, mechanism and situational factors surrounding concussion in a variety of sports.12, 17, 18 While video analysis potentially offers a rich source of information, clear definitions and methodology are required for reviewing such video footage if the review process is to provide valid and reliable data. Recently, Hutchinson and colleagues¹⁶ developed a “heads-up checklist” to code situational factors and injury mechanisms related to concussion in ice hockey. The factors assessed on video analysis were found to have moderate to high inter-rater agreement.¹⁶

Given the variability in game situation between sports, video criteria must be validated for each individual sport. For example, Australian football is a contact sport that is typically played without the use of helmets or other protective equipment (except for mouthguards), and it is played on large grass oval fields, with a 360° direction of play. This is very different from the mechanics of other sports that may be on smaller, rectangular, grass, artificial or ice playing-surfaces, have motion in a limited direction, and/or helmeted players. Each of these variables may modify the video assessment of concussion. Moreover, the rules of the game vary between sports and may provide a unique but effective method of concussion prevention.

Many factors affecting concussion rates in Australian football are potentially modifiable, and require careful assessment. These factors can be assessed with the use of video technology, providing clinicians and administrators a powerful tool to assist with the prevention and management of concussion. The current study used video analysis of concussion in Australian football to investigate the situational factors and mechanism of injury. The aim of the study was to assess the relationship between various player and game-related factors and risk of concussion; and to assess the reliability of video analysis for mechanistic assessment of concussion in Australian football.