Article Text

Download PDFPDF
  1. D O'Sullivan1,
  2. G Fife2,
  3. W Pieter3
  1. 1Chung-Ang University, Anseong, South, Korea
  2. 2Yonsei University, Seoul, South, Korea
  3. 3Keimyung University, Daegu, South, Korea


Background To date, only one study reported the impact attenuation of protective taekwondo headgear. Although Olympic boxing has discontinued the use of headgear, similarities in designs are thought to provide a framework for improvements of taekwondo headgear.

Objective To determine differences in impact attenuation of taekwondo and boxing headgear.

Design Between groups.

Setting Impact biomechanics laboratory.

Participants World Taekwondo Federation (WTF) and Amateur International Boxing Association (AIBA) approved protective headgear.

Risk factor assessment A standardized (ASTM F-2397) martial arts headgear striker was used to impart impacts to a 50th Percentile Male Hybrid III Dummy head and neck complex. All impacts were imparted with a terminal striking velocity of 5.0 m/s (min - max=4.75 −5.15m/s). 2 AIBA-approved headgear (Adidas and Green Hills) and 2 WTF-approved headgear were selected for testing. Each headgear model was fitted to the Hybrid III and impacted a total of 5 times to the left side (temporal) of the headgear along an ASTM-specified impact region. As per ASTM standards low velocity impacts must yield head RLAs below 50g.

Main outcome measurements Resultant linear acceleration (RLA).

Results 1-way ANOVA showed significant differences between the headgear brands (eta2=0.928). The most protection was provided by the taekwondo Adidas headgear (60.5±4.3g) followed by the boxing Adidas (75.4±9.9g), boxing Greenhill (109.4±10.11g), and taekwondo Nike headgear (129.5±8.2g). The greatest difference in RLA was observed between the two taekwondo models (d=−9.54, 95% CI=−13.91 – −5.18), whereas the smallest clinically meaningful difference was observed between the two Adidas models (d=−1.94, 95% CI=−3.20 – −0.29).

Conclusions Each of the headgear brands performed differently under the testing conditions. All models need to be redesigned to mitigate low velocity impacts. Future testing must include impacts at high impact velocities.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.