Article Text

Download PDFPDF

Safety performance evaluation of taekwondo headgear
  1. David M O'Sullivan1,
  2. Gabriel P Fife2,
  3. Willy Pieter3,
  4. Insik Shin4
  1. 1Department of Physical Education, Chung-Ang University, Anseong, Republic of Korea
  2. 2Department of Physical Education, Integrative Sport Sciences Laboratory, Yonsei University, Seoul, Republic of Korea
  3. 3Department of Taekwondo, Keimyung University, Daegu, Republic of Korea
  4. 4Department of Physical Education, Seoul National University, Seoul, Republic of Korea
  1. Correspondence to Gabriel P Fife, Department of Physical Education, Integrative Sport Sciences Laboratory, Yonsei University, 50 Yonsei-Ro, Seodaemungu Shinchondong, Seoul, 120-749 Republic of Korea; fifeg.yonsei{at}


Background With over 20 years of taekwondo concussion research highlighting the high incidence of injury, previous studies recommend an investigation of headgear impact attenuation performance.

Objective To examine impact attenuation differences between the anterior, posterior and sides of selected taekwondo headgear brands.

Design Between-groups.

Setting Biomechanics laboratory.

Methods Five different commercially available taekwondo headgear were selected for impact testing. A 50th percentile Hybrid II Dummy Crash Test head and neck was fitted with the selected helmet and was bolted to a 25 kg steel torso-like structure. Each headgear model was impacted eight times to the anterior, posterior and sides by a 6.75 kg bowling ball at three heights to produce 52.25, 85 and 144 J strikes.

Main outcome measurements Resultant head linear acceleration.

Results Two-way (Helmet×Location) mixed analysis of variance with repeated measures on the second factor was performed to determine the differences between headgear by location of impact. There was a two-way (Helmet×Location) interaction for acceleration (η2=0.368).

Conclusions Taekwondo headgear manufacturers and sport governing bodies must consider improving the design of especially anterior helmet properties.

  • Martial Arts
  • Concussion
  • Head injuries
  • Biomechanics
View Full Text

Statistics from


Taekwondo is a combative Olympic sport with 80 million participants worldwide1 in over 204 countries.2 The concussion incidence in taekwondo is reported to range from 5.5 to 50.2 per 1000 athlete-exposures (A-E).3–5 Siana et al6 first reported on injuries from the 1983 World Taekwondo Championships where two mandibular fractures and one fracture of the zygomatic bone (accompanied with a protruding eye ball) were observed. Although no protective headgear was used during the 1983 World Championships, it must be understood that the area of protection provided by taekwondo headgear does not cover the zygomatic bone. Most recently two deaths7 ,8 were reported during competition with one victim being a 17-year-old novice.8 Also severe subarachnoid haemorrhage caused by the axe kick was recently reported.9

In an effort to improve player safety, protective headgear was first introduced in 1985 and officially used in the 1987 World Taekwondo Federation (WTF) World Championships in Barcelona, Spain.10 However, McIntosh and McCrory11 conclude that headgear in rugby, a similar design to that in taekwondo, does not provide any functional protection against concussion or severe head injury. In 1976, the National Operating Committee on Standards for Athletic Equipment (NOCSAE) in the USA was instituted.12 After its inclusion, and the required safety approval by NOCSAE for all American football (FB) helmets, fatalities in FB decreased 74%.12 Similarly, the American Society for Testing and Materials (ASTM F-2397) impact attenuation testing standard for martial arts headgear was established in 2004.13 However, to our knowledge, no taekwondo headgear adheres to or is approved by ASTM and the head injury incidence in taekwondo continues to rise.14 Therefore, in an effort to better understand the contribution of taekwondo headgear to prevent head injury, the purpose of this study was to evaluate the impact attenuation performance of commercially available taekwondo helmets approved by the Korean Taekwondo Association (KTA) and the WTF.


As the current study was subject to financial limitations, we adopted a testing methodology in a manner similar to that by Moffitt and Lieu,15 which was then later adopted as the ASTM standard for testing protective headgear used in martial arts.13 Where the ASTM methodology calls for the use of a spring-loaded rotating aluminium tube (4.5±0.5 kg), our study employed the use of a pendulum set-up (with confirmed impact reliability) as explained below. It is important to consider that although the testing apparatus used in the current study was different, ensuring reliable impact energies were imparted to the headgear was accomplished and verified through a correlation analysis presented later in this section.

Five different brands (large size) of commercially available KTA-approved and WTF-approved (figure 1) taekwondo headgear (figure 2A–E) were selected for impact testing. To the best of our knowledge, the WTF and KTA do not adhere to an internationally recognised safety standard such as the ASTM F-2397 and it is unclear which criteria, if any, were used to determine approval. To measure the impact attenuation of the headgear, a 500 G tri-axial piezoelectric accelerometer (PCB Piezotronics 356A66, New York, USA) was mounted at the centre of gravity of a Hybrid II dummy head (HCOG). The Hybrid II head and neck was then mounted to a free-hanging iron pendulum (approximately 25 kg) to represent the inertial properties of the upper body of a 50th percentile male hybrid crash test dummy. A 6.75 kg bowling ball was then suspended by a steel-lined cable to impart various impacts to the head-form. To calculate the final velocity of the striker before impact, two light sensors were positioned at a fixed distance before the point of impact (figure 3). Power was supplied to the accelerometer by a signal conditioner (PCB 480B21, New York, USA) and a customised written LabVIEW (version 2010) programme was used for acquisition of light sensor and acceleration data to calculate the preimpact velocity of the striker.

Figure 1

World Taekwondo Federation/Korea Taekwondo Association-approved headgear sticker.

Figure 2

Five commercially available World Taekwondo Federation and Korea Taekwondo Association-approved headgear tested for impact attenuation performance.

Figure 3

Hybrid II dummy head-form and free-hanging iron pendulum with pendulum striker.

Testing procedure

Three different energy impact levels (low, medium and high) were used to investigate the response of the headgear. The high-energy impact was established by the ASTM standard13 and is calculated by an estimation of the foot velocity for a kick to the head (approximately 12 m/s) and the effective mass of the foot (2 kg). For high-energy impact testing (144 J), the striking pendulum (ie, the bowling ball) was raised to a relative height of 2.40 and 1.5 m for medium (85 J) energy impacts. For the low-energy impacts (56.25 J), the striking pendulum was raised to a height of 1 m relative to the intended target. All conditions were repeated at four different locations on the headgear: front, back, left and right sides. For the low-energy and medium-energy impacts, three trials were conducted at each site and two trials for high-energy impacts to ensure products were not destroyed. Testing each headgear started with the lowest impact and then incrementally increased to the medium and high levels. After each impact and a 90 s time delay,13 the headgear was inspected for any structural damage to ensure the integrity of the product was not compromised.

Reliability testing

The reliability of the testing system was confirmed by imparting impacts at four ball drop heights to all four designated sites (frontal, right, left and occipital) over three repetitions for each condition. Pearson correlation coefficients (frontal r=0.724, back r=0.927, side r=0.953) between the striker's pre-impact velocity and the resultant head linear acceleration (RLA) were calculated and deemed to have a strong relationship.16 The reliability of the testing procedures to impart consistent impacts at the specified energy levels was important as our methodology was adopted from the ASTM procedures for testing martial arts safety headgear. Although it is expected that the momentum imparted by the pendulum to the head-form may be different to that of the ASTM recommendations, the consistency with which the energies were matched is thought to be most relevant when considering injury measures as variables used to determine injury potential and helmet impact attenuation are based on imparted energy11 and resultant acceleration.17

Data processing

The maximum acceleration values were calculated and recorded by a customised written LabVIEW2010 (National Instruments, Austin, Texas, USA) programme and the accelerations were saved into text file format. The acceleration data were filtered and processed in accordance with SAE J211-1 channel frequency class 1000.18

Evaluation criteria

Peak RLA is the maximum magnitude of the accelerations from the x, y and z axes measured by the tri-axial accelerometer mounted at the HCOG.17 A variation of the 2004 ASTM F-2397 guidelines was followed and pass/fail criteria were applied to all headgear testing. ASTM F-2397 indicates two levels of impacts (50 and 150 g), and in the event that any headgear model surpasses these impacts, a grade of ‘fail’ was applied. Through personal correspondence with the team (Dennis K Lieu and colleagues), which designed the ASTM martial arts impact attenuation standard, it was explained that the 50 and 150 g impacts were selected as they corresponded to biomechanical studies which identified taekwondo kicks to be able to potentially produce accelerations to the head of those magnitudes.

Statistical analysis

Using STATISTICA V.10 (StatSoft, Tulsa, Oklahoma, USA) a two-way (Helmet×Location) mixed analysis of variance with repeated measures on the second factor was performed to determine the differences between helmets by location of impact. An effect size of 0.2 was considered clinically meaningful.19


Table 1 shows the means and SDs of linear acceleration by helmet and location. There was a two-way (Helmet×Location) interaction for acceleration (η2=0.368). The simple effects analysis indicated that almost all pairwise comparisons were significantly different from each other. By way of summary, figure 4 depicts the results of the linear accelerations collapsed over helmet location in relation to the ASTM standards.

Table 1

Descriptive statistics of accelerations (g) by helmet and location

Figure 4

Impact attenuation performance of the headgear brands. The lower (50 g) and higher (150 g) lines represent the low-energy and high-energy limits set out by the American Society for Testing and Materials (ASTM F2397-04) standards, respectively.


The current study evaluated the impact attenuation performance of commercially available WTF/KTA-approved taekwondo headgear. Recent ASTM impact attenuation standards provide a pass/fail criterion for low-energy (50 g) and high-energy (150 g) head impacts. Although the analysis was done collapsed (ie, combining and averaging all impacts) over impact, perusal of the data seems to suggest that all headgear tested failed either or both low-energy and high-energy impacts (see figure 4). However, an important aspect to note is the difference between the density and Young's modulus of the aluminium impactor used for the ASTM standard, whereas this study used a polyurethane bowling ball. This difference is important as it would affect the coefficient of restitution between the impactor and the headgear. Additionally, magnitude differences were identified for each location (ie, front, rear, left and right) across all helmets, which may represent influences of headgear property characteristics (eg, stiffness and thickness) or even differences due to the anatomical shape of the Hybrid II head. The latter is thought to be the more influential culprit of differences in impact as one tri-axial accelerometer was mounted to the head centre of gravity and the moment of inertia from one location (right side) to another (front) may produce variance in RLA.11 In a 2000 study,20 differences in impact when testing rugby headgear were observed and attributed to head geometry (ie, head form curvature at the centre-front area) especially at the centre-front compared with lateral-side impacts.

Overall mean RLAs resulting from frontal impacts were higher than the back of the headgear. The results of the post- hoc analysis highlight the significant differences between the front and the back of headgear A and E. Koh and associates3 ,4 reported that the two most common kicks that cause concussion are the turning kick and axe kicks. Because the axe kick is typically aimed at the front of the head, and reported to cause severe brain injury, that is, brain haemorrhage,9 the failure to attenuate forces at this location must be considered in future headgear designs. There were no differences between the left and right sides of the helmet. However, as the turning kick is reported to produce the highest RLA (ie, greater than 130 g),21 ,22 improved protection at both sides should be provided.

The first study investigating the various mechanical responses of the skull under loads was carried out on cadavers in 1874.23 Since then, it was reported that the occipital region of the skull is the weakest, followed by the mid-frontal, posterior parietal and anterior interparietal areas.24 Gurdjian and Webster24 attributed these variances to the geometrical characteristics, such as skull shape and thickness. A recent study utilising mathematical modelling and finite-element analysis also confirm that cortical skull thickness and density to be an important biological factor to the determination of skull fracture.25 Furthermore, it was demonstrated that lower forces were required for skull fracture in the frontal region.26 Additionally, the fracture forces for women (3123±623 N) were lower than those of the men (3944±1287 N). With these factors, that is, gender, age and impact site, affecting the strength of the skull, future designs of headgear should consider varying thicknesses around the ‘weaker’ areas (such as the frontal region) to add more protection.

The original implementation of athletic helmet impact attenuation performance testing aimed at reducing the incidence and severity of catastrophic head and neck injury.27 Based on cadaver studies, skull fracture was observed at head accelerations greater than 300 g.28 ,29 To this day, these same pass/fall values are used for helmet impact attenuation testing in an effort to ensure all market headgear in sports, such as football, ice hockey and lacrosse, help prevent catastrophic injuries.17 Recent medical studies continue to highlight the detrimental effects of repetitive concussive impacts and subsequent presentation of chronic traumatic encephalopathy (CTE) in professional athletes as well as high-school football players.30 Although outcomes such as death and CTE represent catastrophic results, efforts to also prevent and reduce the incidence of lower impacts observed in concussion should be of equal concern as they are now understood to possibly lead to CTE.30

The ASTM standards represent a lower range that may be experienced in sport from concussive impacts. However, the upper limits of severe brain injury are not included. Taekwondo peak kick velocities have been measured to range from 13 to 18 m/s depending on the kicking technique,31–33 demonstrating the potential of substantially high RLA values. Linear accelerations of 130 g and higher from the taekwondo turning kick were recently reported.22 It is thought that headgear impact attenuation standards, for not only taekwondo but all contact sports, should aim to protect athletes from the lower concussive forces as well as those exceeding 300 g that are suggested to be life threatening.28 ,29

A number of other impact attenuation testing standards, including those of the ASTM, require the cessation of headgear use by the consumer after a specified time. To this end it is understood that the effects of repetitive blows will affect the ability of the protective helmet to safely attenuate injurious collisions. Our study tested only new headgear and did not investigate the phenomenon of material fatigue and customary wear and tear, which was demonstrated to reduce the ability of the material to attenuate impact forces.20

With some 25 years of research in taekwondo highlighting a high incidence of concussion, as well as catastrophic head injuries, not only does the WTF hold the utmost responsibility for ensuring player safety with regard to improved headgear designs, but a concentrated effort to enact the first injury surveillance system for this sport is imperative. Along with the catastrophic head injuries are recent confirmations of the manifestation of CTE30 in other contact sports due to repetitive head trauma, which should be a future area of study within this cohort. A well-accepted approach to prevent injury by Van Mechelen,34 which is explained in depth elsewhere, is suggested. A recent review of taekwondo competition injuries provided a number of injury preventative measures, such as improved coach and referee education of injury recognition, preparticipation examination and improved medical care at events, protective equipment, improved defensive manoeuvres and rule changes.35 Specifically for rules, an effort to discourage athletes from focusing kicks to the head by decreasing the allotment of points awarded to the head from four to one point is paramount, as recent research demonstrated increased head kicks36 and possible head injuries.14


Although the current study provides the first known assessment of taekwondo headgear impact attenuation, limitations when using anthropometric test devices (ATD) (eg, Hybrid II) should be addressed. The design of the Hybrid II structure does not permit realistic neuromuscular responses relating to the neck, particularly as there are no pretuning capabilities prior to impact, which may often be the case in real taekwondo bouts. Early investigations on the Hybrid II and III biofidelity reported nearly three to five times greater neck stiffness for the Hybrid III compared to the human neck, and poor response changes in velocity.37–39 On the other hand, the Hybrid II has also been observed to have minimised energy absorption capabilities as well as a less biofidelic head,40 which are attributed to a reduction in the approximation of head injury.29 A recently designed ATD (ie, THOR Advanced Crash Test Dummy) is currently used by the National Highway Traffic Safety Administration (USA) due to its significantly improved biofidelity compared to the earlier models designed in the 1960s and 1970s.41 However, the current study used a more economical alternative. The issue of ATD biofidelity indicates the need for caution when applying the results from laboratory studies.


To the best of our knowledge this is the first study to evaluate the impact attenuation performance of taekwondo headgear that is approved for competition use by the WTF and KTA. Although the ASTM standards for martial arts headgear impact attenuation testing were instituted in 2004, the WTF/KTA-approved helmets that were used in the current investigation do not pass these criteria. Currently, at WTF-sanctioned competitions, all headgear must be officially approved by the WTF. Based on the current investigation, it is not clear which criteria, if any, were used to warrant the official approval stamp imprinted on these helmets. Alarmingly, and to the best of our knowledge, even reports of deaths7 ,8 during full-contact taekwondo competitions have not initiated any action by governing bodies to improve the impact attenuation performance of approved headgears. After an improved safer version is designed, it is recommended that the WTF adopt the standard specification for protective headgear used in martial arts (ASTM F2397-04). Adherence to these standards for all WTF-approved headgear may better ensure safety of taekwondo athletes and help avoid further catastrophic events during competition. Furthermore, it may do well to amend the current ASTM standards by including impact strikes exceeding 150 g to better represent the more acute severe injuries, such as skull fractures. Although current headgear and helmet testing methods do not employ the assessment of rotational accelerations, future investigations are encouraged to employ the use of oblique impact testing as recommended by Aare et al.42

What are the new findings?

  • The first known laboratory study to assess impact attenuation performance of taekwondo safety headgear.

  • This investigation reports that selected helmets approved by the World Taekwondo Federation and Korea Taekwondo Association do not pass low-impact and high-impact trials.

How might it impact on clinical practice in the future?

  • Provides clinicians with scientific evidence of taekwondo headgear performance.

  • Taekwondo headgear manufacturers may take recommendations to improve impact attenuation performance that adhere to American Society for Testing and Materials standards.

  • Clinicians will be able to make an objective decision if headgear provides sufficient protection during competition.

  • Suggestions for amending current martial arts headgear impact attenuation testing procedures to include impacts that might cause concussion and catastrophic head injury.


We would like to thank Hyunju Lee for editing the images included in this article and Humanetics ATD for supplying the Hybrid II for this study.


View Abstract


  • Contributors DO contributed to conception and design; acquisition, analysis and interpretation of the data, drafting and critical revision, and final approval of the article. GPF contributed to conception and design; acquisition and interpretation of the data, drafting and critical revision, and final approval of the article. WP contributed to conception and design; acquisition, analysis and interpretation of the data, drafting and critical revision and final approval of the article. IS contributed to conception and design; interpretation of the data, drafting and critical revision and final approval of the article.

  • Funding This study was supported by the Yonsei University Research Fund.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • ▸ References to this paper are available online at

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.