Injury Rates and Risk Factors in Competitive
Tsharni R. Zazryn, BAppSc(Hons),* Paul R. McCrory, MBBS, PhD,†
and Peter A. Cameron, MBBS, MD‡
Objective: To determine injury rates and risk factors for injury in
a cohort of professional boxers.
Design: Retrospective cohort design reporting on data collected for
a ﬁght statistics database maintained by the Professional Boxing and
Combat Sports Board of Victoria, Australia. Data were extracted for
the years January 1997 through June 2005.
Setting: Victoria, Australia.
Participants: 545 professional boxers (age, 18 to 43 years) who
participated in a total of 907 ﬁghts over the study period.
Assessment of Risk Factors: Independent variables under
investigation included age, gender, weight, bout exposure, and
location of the bout (within or outside of the State of Victoria).
Main Outcome Measures: Physician-reported acute boxing
injuries occurring during bouts of any region or nature.
Results: 214 injuries were sustained over the 8.5 years, correspond-
ing to an injury rate of 23.6 per 100 professional ﬁghts. The majority
of these injuries were lacerations to the head and face. An increasing
age and an increasing number of ﬁghts were both signiﬁcant
predictors of injury.
Conclusions: Injury reduction strategies for professional boxing
might include restrictions of eligibility to ﬁght based on age and
boxing bout exposure. Future research using prospective cohort
designs and standardized injury deﬁnitions are needed to conﬁrm
these results. Greater mechanistic detail and more complete data
entry are necessary to ensure that optimal injury prevention strategies
can be developed and implemented. Upon conﬁrmation of the results
of this study, the Professional Boxing and Combat Sports Board
of Victoria may consider different criteria upon which to sanction
Key Words: boxing, bout, professional
(Clin J Sport Med 2009;19:20–25)
Limited epidemiological research has been carried out in
professional boxing. Instead, much of the boxing literature has
focused on the amateur discipline of this sport. As it is
unknown whether the data collected in the amateur discipline
is applicable to the professional cohorts, it is important that
accurate and valid data for professional boxers are obtained.
A general injury proﬁle has emerged from the limited
number of international studies conducted involving pro-
fessional boxers. The head is the most commonly injured body
region, followed by the hands and ﬁngers, and the majority of
the injuries are bruises and lacerations.
More severe injuries,
such as concussions and fractures, are reported less frequently.
Unfortunately, little is known about injury risk factors for these
participants. Thus, data relating to the rates of injury and the
risk factors for injury development are needed before injury
prevention strategies can be developed and successfully
In professional boxing, only 5 studies have been
published that report any injury rates, and no study has
addressed injury risk factors for this population in the past 30
Rather, what is known about injury risk factors in this
sport has arisen primarily from anecdotal beliefs about
potential injury risks.
There is a clear need for prospective
studies aimed at determining injury rates and analyzing risk
factors associated with contemporary professional boxing.
The aim of this study was to analyze the injury data
collected during competition by a professional boxing
association to calculate injury rates and determine risk factors
for professional boxing.
This study employed a retrospective cohort design using
a ﬁght statistics database maintained by the Professional
Boxing and Combat Sports Board of Victoria (PBCSBV),
Australia. This database is used to collect data on ﬁght
outcomes for all professional boxers registered in Victoria, and
for any professional boxers who come to Victoria to ﬁght.
All boxers who were registered with the PBCSBV and
had a ﬁght (regardless of the ﬁght location) or those registered
Submitted for publication January 29, 2008; accepted September 12, 2008.
From the *Departments of Health Science and ‡Epidemiology and Preventive
Medicine, Monash University, Frankston, Victoria, Australia; and †Centre
for Health, Exercise and Sports Medicine and the Brain Research Institute,
University of Melbourne, Melbourne, Australia.
Tsharni Zazryn was supported by a National Health and Medical Research
Council Postgraduate Scholarship.
The authors state that they have no ﬁnancial interest in the products mentioned
within this article.
Reprints: Tsharni Zazryn, BAppSc(Hons), Department of Health Science,
Monash University, PO Box 527, Frankston, Victoria 3199, Australia
Copyright Ó2009 by Lippincott Williams & Wilkins
20 Clin J Sport Med Volume 19, Number 1, January 2009
with other Boards but ﬁghting within the State of Victoria for
the 8.5-year period from January 1997 to June 2005 comprised
the study cohort.
Instruments and Procedures
The PBCSBV dataset is maintained within a Microsoft
Access database. Data for the appropriate period were
extracted from the Access database and transferred into
Microsoft Excel 2002. Each ﬁght recorded in the database
potentially had 1 or 2 records.
Fights with 1 record in the
database indicated that a Victorian-registered boxer was
ﬁghting against a non-Victorian boxer. Where 2 records for
the 1 ﬁght were recorded, this indicated either 2 Victorian-
registered boxers ﬁghting each other in any location or any
ﬁght in the State of Victoria. In order for accurate injury rates
to be calculated, 2 ﬁght records (1 for each boxer) for every
ﬁght in the database had to exist. A manual search of the
database for each ﬁght was therefore undertaken. Where 2
records for a bout had not been entered into the database, the
missing data were obtained from the paper-based records of
the PBCSBV and added to the database. Conversely, where
multiple records (ie, more than 2) had been entered for 1 bout,
the most complete record was kept and the other(s) deleted.
Potential information entered into the ﬁght statistics
database for each bout included the boxer’s name, the
opponent’s name, the date of the ﬁght, weight at weigh-in
24 hours before the contest, the number of scheduled rounds,
the number of completed rounds, the ﬁght result (win, loss, or
draw), the ﬁght result status [decision determined by points,
knock-out (KO) or technical knock-out (TKO)], the location of
the ﬁght (entered as the name of the venue), and any sustained
injuries and any recommendations made by the ringside
physician regarding when the boxer could next ﬁght. This was
linked to the table of personal information for each boxer,
which allowed the date of birth and gender of the boxer to be
determined. Not all of these variables were entered fully, and
thus not all could be used to determine risk factors for injury.
The possible risk factors for injury that could be
determined from the dataset included age at the time of the
ﬁght (determined by using the date of birth of the boxer, which
is entered into the database at registration, and the date of
the ﬁght), gender, weight, the number of ﬁghts (determined
by the number of ﬁght records for each boxer in the dataset,
which were ordered from earliest to most current in date), and
location of the ﬁght (as Victorian or non-Victorian, recorded
on the basis of venue).
At all Victorian bouts a ringside physician is required to
undertake pre-bout and post-bout medical checks of partic-
ipants. After each bout, the ringside physician notes any
injuries that were either reported to them by the boxers or
observed/treated by the physician. Therefore, the injury
deﬁnition used in this study was any injury reported to or
by a ringside physician after bout participation. The ﬁght detail
form provides 3 lines of space for Any Injuries, and the exact
detail provided by each physician was not standardized. Any
injuries (either physician or boxer reported) will be referred to
as reported injuries in the data below. The injury details were
entered into the Access database in the form of a text narrative
that was copied verbatim from the completed form. For the
8.5-year period of the current data collection, 3 different
doctors were serving as ringside physicians within Victoria.
For Victorian-registrants who fought interstate, they were
required to have a form completed by the appropriate boxing
authority. The form detailed the ﬁght result and any injuries
sustained and had to be returned to the PBCSBV within 1
week of the bout.
Injury details were recoded from the text narrative
according to the guidelines provided by the Australian Sports
Injury Data Dictionary.
Only details pertaining to the region
and nature of injury were available for coding. No causes of
injury data were included in the ringside physician notes.
The data were transferred into both the Statistical
Package for the Social Sciences (SPSS) version 12.0 and
Intercooled STATA version 9.2 for analysis. A competitive
bout was deﬁned by 2 ﬁght records (ie, both the boxer and
the opponent were included in the analysis). Basic analyses
were conducted for all questions with descriptive statistics
used as appropriate to provide demographic and injury details.
Exposure was quantiﬁed in terms of the number of ﬁghts
participated in during the study period for each boxer. Injury
rates were calculated in a number of different ways that are
consistent with the previous literature.
This included injury
rates based on the reported injuries only and injury rates based
on the reported injuries and any ﬁghts in which a boxer lost by
TKO or KO. One previous amateur boxing study has stated
that, if a ﬁght has been terminated as a result of a KO or the
referee stopping the contest (RSC in amateurs, similar to
a TKO in professional boxing), the boxer is assumed to have
had evidence of some neurological dysfunction at that time.
As such, an injury event is assumed to have occurred.
However, study results have shown conﬂicting evidence of
neurological dysfunction when comparing baseline neuro-
psychological test results to those obtained after a KO or TKO
Thus, injury rates were calculated using both
reported injuries and situations in which a possible injury may
have occurred (based on a KO or TKO loss).
Univariate and multivariate logistic regressions were
used to identify predictors of injury (for reported injuries
only). Bouts in which a KO or TKO was recorded as the ﬁght
result but no injury was reported were not included as injuries
for the risk factor analysis. For each model, odds ratios, 95%
conﬁdence intervals, and Pvalues were calculated. In analyz-
ing all statistical tests, a signiﬁcance level of alpha 0.05 was
used. The clustering effect of repeat injuries to a boxer across
different bouts was accounted for by using robust standard
errors in the multivariate logistic regression model.
A total of 907 professional boxing bouts were
participated in by 545 Victorian-registered boxers or within
the State of Victoria for the 8.5-year period. The majority of
the ﬁght participants were male (98.3%), with an average age
at the time of a ﬁght of 27.9 years (range, 18.1 to 43.6 years).
The average age of boxers at their ﬁrst professional ﬁght
was 27.7 years (SD, 5.1 years; range, 18.1 to 43.3 years). The
q2009 Lippincott Williams & Wilkins 21
Clin J Sport Med Volume 19, Number 1, January 2009 Injury and Risk in Professional Boxing
percentage of ﬁghts participated in per person is shown in
Table 1. Over three–quarters of boxers (75.2%) had 3 ﬁghts or
fewer over the time period.
Injuries were reported in 177 (19.5%) of the 907 ﬁghts.
As multiple injury events occurred in some ﬁghts (range, 1 to 3
injuries), the total number of reported injuries was 214. This
corresponds to a reported injury rate of 23.6 per 100
professional ﬁghts (Table 2). When KO and TKO losses were
included as an injury, the injury rate increased to 60.7 per 100
The majority of injuries were open wounds/lacerations
to the head region (61.7%), followed by concussions (11.7%)
and fractures, primarily involving the hands/ﬁngers and nose
(8.4%). Table 3 provides details of the regions and natures of
the injuries sustained by this cohort.
One death during the study period also occurred. The
boxer was participating in his third professional ﬁght, but was
registered in another Australian state. This boxer collapsed
during the sixth round of an eight-round ﬁght and sub-
sequently died from complications of a left frontoparietal
subdural hemorrhage with cerebral infarction and associated
oedema. A coronial inquest focused on the issues of weight
loss, dehydration, ambulance delay, and mismatching of the
opponent as factors contributing to the death.
presented in the inquest did not support any of these factors as
ultimately contributing to the death of the boxer.
Injured boxers (whether 1 or more injuries) were
signiﬁcantly more likely to lose their ﬁghts than noninjured
boxers (OR, 2.1; 95% CI, 1.5 to 3.0). Further, injured boxers
were signiﬁcantly more likely to ﬁnish their ﬁght before the
scheduled number of rounds (OR, 1.4; 95% CI, 1.01 to 1.92).
As shown in Table 4, independent predictors of injury
included the age of participants at the time of the ﬁght (with
older boxers more likely to be injured; OR, 1.04; 95% CI, 1.01
to 1.08) and the number of ﬁghts each boxer had participated
in (with boxers with a greater number of ﬁghts having an
increased risk of injury; OR, 1.07; 95% CI, 1.05 to 1.10).
This increasing risk of injury as ﬁght numbers increased is
depicted in Figure 1. As age increased, a small but consistent
increase in injury risk was seen as demonstrated in Figure 2.
Even after adjusting for other variables in a multivariate
logistic regression and the potential for clustering due to some
boxers having injury events in multiple ﬁghts, an increasing
number of ﬁghts remained a signiﬁcant predictor of injury
Almost one–ﬁfth of professional boxing ﬁghts resulted
in a reported injury in this study. Only 3 other studies have
reported injury rates per 100 ﬁghts in boxing.
Only 1 of
these reported on injuries occurring to professionals, and that
study was carried out over a 1-year period with 14 boxers from
the same cohort that this current study used. The injury rates
were slightly higher in the 1-year study (33.3 injuries per 100
A greater focus on injuries within the 14 boxers in the
cohort study, as well as fewer ﬁghts being detailed over that
time, may have led to an increased injury rate within that
group. All other studies reporting injury rates per 100 ﬁghts
have been undertaken in children or amateur competitors.
Similar injury rates for the professional boxers in this study
have previously been reported for amateur boxers (15.5, 22.8,
and 25.0 injuries per 100 ﬁghts).
When considering reported injuries as well as losses by
KO or TKO, the injury rate was 3 times that calculated for
reported injuries alone, with approximately two–thirds of
ﬁghts resulting in an injury. If some form of neurological
dysfunction has in fact occurred to participants who lose a
ﬁght by KO or TKO, it may be that a number of injuries are
remaining undiagnosed, untreated, and unreported in pro-
fessional boxing. A more detailed understanding of the
neurological consequences of KOs or TKOs in boxers is
needed to determine if some form of treatment is required for
this group after a ﬁght.
The risk factors of age, weight, gender, the number of
ﬁghts participated in, and the location of the ﬁghts were able to
be assessed from the database. Current age, age at beginning
boxing, and age at retirement from boxing have all long been
considered potential risk factors for injury.
This study was
able to assess current age as a potential risk factor and found
TABLE 1. Number of Fights Participated in Per Person From
1997–2005 (n = 545 Boxers)
Fights Per Person
1 292 53.6
2 72 66.8
3 46 75.2
4 30 80.7
5 17 83.8
6 13 86.2
7 13 88.6
8 8 90.1
9 6 91.2
10 0 91.2
$11 48 100.0
TABLE 2. Injury Rates in Professional Boxing Over a
Injury Rate Per
Losses by KO/KO
1997 90 17 18.9 52.2
1998 70 17 24.3 50.0
1999 85 25 29.4 68.2
2000 88 29 33.0 73.9
2001 112 32 28.6 59.8
2002 104 27 26.0 62.5
2003 139 35 25.2 65.5
2004 135 25 18.5 62.7
2005* 84 7 8.3 51.2
Total 907 214 23.6 60.7
*Until end June 2005.
22 q2009 Lippincott Williams & Wilkins
Zazryn et al Clin J Sport Med Volume 19, Number 1, January 2009
that increasing age at the time of the ﬁght was an independent
signiﬁcant predictor of injury, and it remained such when
adjusting for other predictor variables but not when adjusting
for clustering. While a speciﬁc age at which an increased risk
becomes apparent was not found in the data, a steady increased
risk for each year of life was found.
While not a modiﬁable injury risk factor, competitor age
is used to restrict participation in professional boxing, with no
participant allowed to compete if they are younger than 18
years of age at their ﬁrst ﬁght (they may start at any point after
18 years). Other than that, no age restrictions exist in
professional boxing. The data have shown that there is a
potential for increased injury risk between the age of 18 and
about 23 years of age. The level of risk then plateaus
until approximately 28 years of age when an increased risk is
again apparent until about 35 years of age.
The exposure experienced by professional boxers (in
terms of number of ﬁghts) has previously been reported as
a potential injury risk factor.
This study has shown that an
increasing number of ﬁghts does signiﬁcantly increase the risk
of acute boxing injury (even once all other factors are
accounted for). In particular, when ﬁghts are categorized into
5-ﬁght groupings, participants who have had between 11 and
15 ﬁghts and 16 or more ﬁghts are at almost a 2-fold increased
risk of injury compared to those who have had up to 5 ﬁghts.
This study has also shown, however, that this level of exposure
TABLE 3. Regions and Natures of Professional Boxing Injuries Over an 8.5-Year Period (n = 214 Injuries)
Laceration/Cut Strain Rupture
Cheek 3.3 0.5 3.7
Eye (NFS) 0.9 1.4 27.1 0.5 29.9
Eyebrow 0.5 17.3 17.8
Eyelid 0.5 5.6 6.1
Ears 0.5 0.9 0.5 1.9
Face (NFS) 0.5 0.5
Intracranial 11.7 11.7
Jaw/chin 0.5 0.5 0.5 0.5 1.9
Mouth 0.9 0.9
Neck 0.5 0.5
Nose 0.5 2.3 0.5 3.3
Scalp 0.5 5.6 6.1
Teeth/gums 0.9 0.9
Tongue 0.5 0.5
Head (NFS) 0.5 0.5
Hands/ﬁngers 4.7 0.5 1.9 7.0
Shoulder 0.5 0.5
Arm (NFS) 0.5 0.5
Internal 0.9 0.9
Ribs 0.5 0.5 0.9
Ankle 0.5 0.5 0.9
Leg (NFS) 0.5 0.5
Not Speciﬁed 0.5 2.3 2.8
Total 1.4 4.7 11.7 0.5 8.4 1.4 64.0 0.9 0.9 0.5 5.6 100.0
Percentages do not sum to 100% due to rounding.
TABLE 4. Independent Predictors of Injury for
Predictor Variable n Odds Ratio (95% CI) P
Age 1707 1.04 (1.01 to 1.08) 0.014
Weight 1206 1.00 (0.98 to 1.02) 0.880
Male gender 1813 0.63 (0.15 to 2.68) 0.534
Number of ﬁghts 1813 1.07 (1.05 to 1.10) ,0.001
1 to 5 (reference)
6 to 10 1.81 (1.22 to 2.68) 0.003
11 to 15 2.48 (1.51 to 4.08) ,0.001
16+ 3.38 (2.07 to 5.53) ,0.001
Location of ﬁght
Victorian (reference) 857
Non-Victorian 547 0.77 (0.55 to 1.08) 0.129
q2009 Lippincott Williams & Wilkins 23
Clin J Sport Med Volume 19, Number 1, January 2009 Injury and Risk in Professional Boxing
is potentially rare, as only 8.8% of boxers in this cohort had
more than 10 ﬁghts in the Victorian jurisdiction. However, as
discussed in the study limitations below, it is known that some
skewing of the data towards a lower number of ﬁghts is
apparent in this database as a result of interstate or
international competitors only ﬁghting 1 or 2 times within
the Victorian jurisdiction.
Although this study has been unique in the provision of
injury rates and risk factors for professional boxing, it has been
limited by the use of a database that was not originally
designed for injury surveillance purposes. Thus, the retro-
spective cohort design using a database designed to maintain
ﬁght statistics has hindered the ability to fully explore the
breadth of potential risk factors for boxing injury. Future
iterations of this database should include mechanistic data (in
the form of detailed text narratives), complete entry of the
weight divisions being participated in, complete entry of
the number of rounds scheduled and actually completed in the
bout, and reporting on the use of protective equipment types
and styles in order for it to be adequate for injury prevention
purposes. Further, although only three physicians reported on
the ﬁghts detailed in the database, it is important for the
accuracy and integrity of the data that a standardized injury
deﬁnition be implemented so that the same information is
being collected regardless of time of collection and identity of
This database is also limited in that it provides detail of
only one aspect of boxing exposure in the one state of
Australia and does not detail exposure related to amateur ﬁghts
or training times. Finally, boxers who were not Victorian
registrants were included in this dataset; therefore, the
exposure may be skewed towards a lower number of ﬁghts,
as their exposure in other states was unknown. Consideration
needs to be given to having a national surveillance system that
documents both amateur and professional careers and that
measures training settings as well as competition. Such
a database would allow better long-term follow-up of boxers
and would avoid issues regarding the exposure of registrants
ﬁghting in other states.
Future research should aim to conﬁrm the results of this
study and document other potential risk factors for boxing
injury. Factors such as exposure time (based on the number of
minutes of completed rounds), the exposure experienced
outside of Victoria for non-Victorian registrants, amateur
career exposure, the time between each bout for each boxer,
and some measure of skill (including ﬁghting style and
matching of opponents) should be studied.
the results could be used by the PBCSBV to change current
practice if required. At present, boxers who register with the
Board must get clearance from a general practitioner on an
annual basis to be allowed to ﬁght and must have yearly
serology testing. At registration, these boxers are also required
to have a brain magnetic resonance imaging (MRI) scan, with
this scan then repeated at 3-year intervals regardless of the age
or experience of the boxer. The PBCSBV must sanction each
ﬁght, with promoters providing the Board with proposed ﬁght
FIGURE 2. Smoothed (log transformed) risk of injury with
TABLE 5. Multivariate Analysis of Risk Factors for Professional
Boxing Injuries (n = 1364 Fight Participations)
Predictor Variable Odds Ratio [95% CI] P
Adjusted for All Other Predictor Variables
Age 1.04 (1.01 to 1.08) 0.019
Male gender 0.60 (0.14 to 2.59) 0.497
Number of ﬁghts 1.04 (1.02 to 1.07) ,0.001
Non-Victorian ﬁght location 0.78 (0.55 to 1.09) 0.148
Adjusted for All Other Predictor Variables and Potential Clustering of
Multiple Injuries in Multiple Bouts for Boxers
Age 1.04 (1.00 to 1.08) 0.051
Male gender 0.67 (0.20 to 2.29) 0.524
Number of ﬁghts 1.04 (1.01 to 1.08) 0.006
Non-Victorian ﬁght location 0.77 (0.55 to 1.07) 0.122
*Standard error adjusted for 389 clusters.
FIGURE 1. Increasing injury risk (as determined by odds ratios)
as ﬁght participation numbers increase.
24 q2009 Lippincott Williams & Wilkins
Zazryn et al Clin J Sport Med Volume 19, Number 1, January 2009
combinations a minimum of 1 month from the proposed date
of the ﬁght. Upon conﬁrmation of the results of this study, the
PBCSBV may consider different criteria upon which to
sanction a ﬁght with regard to a boxer’s age and the number of
bouts they have previously had or any other conﬁrmed risk
factor. Further, greater deliberation of the ﬁght details
(including who the opponent is, the opponents’ age and
experience, and the proposed length of the ﬁght) may be
required in making these decisions.
Additionally, more intensive medical screening and
evaluations of boxers at the start of their professional career
may also be warranted. This screening at registration should
continue to be (at a minimum) a full medical, serology, and
a brain MRI scan to limit the likelihood of boxers ﬁghting
with potential injury risk factors. Neuropsychological testing
could also be performed at this time to provide a baseline
measure for each boxer. This screening should be carried
out annually. For boxers considered at risk of injury on the
basis of their age and experience, more intensive screening
may be warranted.
This study has provided details on injury rates and risk
factors obtained from the analysis of a comprehensive boxing
database in Victoria, Australia. The rates are based on over 900
ﬁghts over an 8.5-year period. Two hundred fourteen injuries
were reported by either physicians or boxers, corresponding to
an overall injury rate of 23.6 per 100 ﬁghts. When ﬁghts lost
by KO or TKO were also included in the injury rate, almost
two–thirds of all professional ﬁghts resulted in an injury
(60.7 per 100 ﬁghts). Injured boxers were more likely to be
older and had participated in more ﬁghts than non-injured
boxers. Greater screening of boxers early on in their careers
and once they have surpassed 5 ﬁghts may be needed to reduce
the potential risk of injury in this group. Further research into
potential risk factors for injuries is needed for effective injury
prevention strategies to be developed and implemented in pro-
1. Jordan B, Campbell E. Acute injuries among professional boxers in
New York: a two-year survey. Phys Sportsmed. 1988;16:87–91.
2. Zazryn T, Finch C, McCrory P. A16-year study of injuries to professional
boxers in the state of Victoria, Australia. Br J Sports Med. 2003;37:
3. Zazryn T, Cameron P, McCrory P. A prospective cohort study of
injury in amateur and professional boxing. Br J Sports Med. 2006;40:
4. Bledsoe G, Li G, Levy F. Injury risk in professional boxing. South Med J.
5. McCown I. Boxing injuries. Am J Surg. 1958;98:509–16.
6. Jordan B. Boxing. In: Epidemiology of Sports Injuries. Caine D, Caine C,
Lindner K, eds. Champaign, IL: Human Kinetics Publishers, Inc; 1996:
7. Scott I, Finch C, Ozanne-Smith J, et al. Counting Injuries Out of Boxing.
Melbourne: Monash University; 2001.
8. Australian Sports Injury Data Working Party. Australian Sports Injury
Data Dictionary: guidelines for injury data collection and classiﬁcation
for the prevention and control of injury in sport and recreation. Canberra:
Sportsafe Australia (Australian Sports Commission) and Sports Medicine
9. Estwanik J, Boitano M, Ari N. Amateur boxing injuries at the 1981
and 1982 USA/ABF national championships. Phys Sportsmed. 1984;12:
10. Porter M, O’Brien M. Incidence and severity of injuries resulting from
amateur boxing in Ireland. Clin J Sport Med. 1996;6:97–101.
11. Critchley G, Mannion S, Meredith C. Injury rates in Shotokan karate. Br J
Sports Med. 1999;33:174–177.
12. Pieter W, Zemper E. Head and neck injuries in young taekwondo athletes.
J Sports Med Phys Fitness. 1999;39:147–153.
13. Koh J, Cassidy JD. Incidence study of head blows and concussions in
competitive taekwondo. Clin J Sport Med. 2004;14:72–79.
14. Butler R, Forsythe W, Beverly D, et al. A prospective controlled
investigation of the cognitive effects of amateur boxing. J Neurol
Neurosurg Psychiatr. 1993;56:1055–1061.
15. Matser E, Kessels A, Lezak M, et al. Acute traumatic brain injury
in amateur boxing. Phys Sportsmed. 2000;28:87–92, 5–6.
16. Moriarity J, Collie A, Olson D, et al. A prospective controlled study of
cognitive function during an amateur boxing tournament. Neurology.
17. West I. Record of investigation into death - Ahmad Zia Popal. Report
No.: 1001/01. Melbourne, Victoria: State Coroners Ofﬁce; 2005
18. Professional Boxing and Combat Sports Board of Victoria. Rules and
scoring for kickboxing and boxing competitors. Victoria: Sport and
Recreation Victoria; 2002.
q2009 Lippincott Williams & Wilkins 25
Clin J Sport Med Volume 19, Number 1, January 2009 Injury and Risk in Professional Boxing