doi:10.1136/bjsm.36.6.446
Br. J. Sports Med. 2002;36;446-451
S Drawer and C W Fuller
football using a risk based assessment process
Evaluating the level of injury in English professional
http://bjsm.bmjjournals.com/cgi/content/full/36/6/446
Updated information and services can be found at:
These include:
References
http://bjsm.bmjjournals.com/cgi/ ... 6/446#otherarticles
11 online articles that cite this article can be accessed at:
http://bjsm.bmjjournals.com/cgi/content/full/36/6/446#BIBL
This article cites 10 articles, 7 of which can be accessed free at:
Rapid responses
http://bjsm.bmjjournals.com/cgi/eletter-submit/36/6/446
You can respond to this article at:
service
Email alerting
top right corner of the article
Receive free email alerts when new articles cite this article - sign up in the box at the
Topic collections
• Sports Medicine (1329 articles)
• Injury (833 articles)
• Other Epidemiology (1667 articles)
Articles on similar topics can be found in the following collections
Notes
http://www.bmjjournals.com/cgi/reprintform
To order reprints of this article go to:
http://www.bmjjournals.com/subscriptions/
To subscribe to British Journal of Sports Medicine go to:
Downloaded from bjsm.bmjjournals.com on 22 October 2006
ORIGINAL ARTICLE
Evaluating the level of injury in English professional
football using a risk based assessment process
S Drawer, C W Fuller
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Br J Sports Med 2002;36:446–451
Objectives: To show how epidemiological data can be presented and analysed in frequency based
and risk based formats and how risk based information can simplify management decisions on injury
prevention strategies in professional football.
Methods: The club physiotherapists at four English professional football clubs prospectively recorded
players’ injuries over the period November 1994 to May 1997. The nature, location, and mechanism
of each injury and the specific numbers of days that players were unavailable to train or play as a result
of injuries were recorded. The rates of injury were evaluated on a risk matrix using the number of days
and the estimated costs of absence as measures of injury consequences.
Results: There was a significant difference in the time lost through injury as a function of injury severity
(p<0.001). Slight and minor injuries accounted for 51% of all injuries but represented only 17% of
the risk from injury, whereas major injuries accounted for only 12% of the number of injuries and 47%
of the risk. Player to player contact injuries accounted for 39% of the risk of injury, and football specific
activities accounted for 47% of the risk. The risks of acute injury in professional football were three
orders of magnitude greater than those in the construction, manufacturing, and service sectors of industry.
Conclusions: The risks associated with minor, moderate, and major acute injuries and osteoarthritis in
lower limb joints of professional footballers were found to be unacceptable when evaluated against
work based risk criteria used by the Health and Safety Executive. All stakeholders within professional
football were shown to have an important contribution to make in reducing the overall level of risk to
players through the provision of risk prevention strategies.
Under UK health and safety legislation, employers are
required to ensure the health, safety, and welfare of
their employees,1 and, in order to comply with this legislation,
they must carry out risk assessments to identify the
level of occupational risk.2 Employees in the United Kingdom
suffer, on average, 0.36 reportable injuries per 100 000
working hours within the range 0.03 for the finance sector to
1.3 for the mining/quarrying sector.3 Footballers, however,
suffer 710 reportable injuries per 100 000 hours of training
and competition,4 which is several orders ofmagnitude greater
than that found for most occupations in the United Kingdom.
In addition to the high rates of injury, 47% of players5 are
forced to retire from professional football as a result of an
acute or chronic injury. Risk taking and aggressive behaviours
by players contribute to 28–33% of the players’ injuries,6 7 and
3.7% of fouls result in minor or moderate injuries.8 Although
the injury rate caused by foul play is high at 500
injuries/100 000 hours, the rate of injury from other causes is
even higher at 750 injuries/100 000 hours.4 The significance of
these high levels of injury is increased by the fact that most
players have only a poor understanding of the beneficial contribution
that prevention strategies can make to reducing the
levels of injury9 and the fact that clubs provide an inadequate
level of support services to players for injury prevention.10
Work based injuries are normally measured using frequency
based performance indicators, such as the number of
injuries per 100 000 hours or per 1000 employees. This method
of measurement, however, does not reflect the true level of risk
because it does not take into account the number of days that
employees are absent from work or the costs incurred by
employees and employers as a result of an injury. It is possible,
using a frequency based assessment, for a large number of
minor injuries to obscure a greater impact from a small
number of major injuries. For this reason, assessments should
review the levels of injury using risk based criteria, such as the
number of days lost by employees or the financial costs or
losses experienced by employees and employers.
In the United Kingdom, the Health and Safety Executive
(HSE) have provided guidelines on the levels of work based
risk that are considered to be negligible, acceptable, tolerable,
and unacceptable. A negligible level of risk11 is “a level of risk,
usually presumed to be below 1 in a million per annum and
perhaps much lower, of seriously adverse consequences occurring,
where no thought is given to their likelihood in the conduct
of normal life”. An acceptable level of risk11 is “a risk in
the region of 1 in a million of a serious adverse occurrence,
where the conduct of life is not affected provided that we are
in fact satisfied that reasonable precautions are in place”. A
tolerable level of risk11 is “a range of risk that we do not regard
as negligible or as something we might ignore, but rather as
something we need to keep under review and reduce it still
further if and as we can”. An unacceptable level of risk11 is “a
risk which is beyond (above) the region of tolerability”. A
serious adverse occurrence in the context of injury is normally
taken to mean a fatality. The HSE12 have defined the region of
tolerable risk to an employee as a probability of between 1 in 5
´ 104 and 1 in 103 of a fatality occurring, and the region of
acceptable risk as a probability of between 1 in 106 and 1 in 5
´ 104 of a fatality occurring.
The first aim of this paper is to assess epidemiological data
of injuries to professional footballers on a risk based format
and to evaluate the results against the HSE guidance on the
acceptability and tolerability of risk to people at work. The
second aim is to show how an assessment of injury data on a
risk based format can assist football clubs in identifying the
significant sources of injury, thereby supporting the development
of injury prevention strategies.
See end of article for
authors’ affiliations
. . . . . . . . . . . . . . . . . . . . . . .
Correspondence to:
Dr Fuller, Scarman Centre,
University of Leicester, 154
Upper New Walk,
Leicester LE1 7QA, UK;
cwf2@le.ac.uk
Accepted 17 April 2002
. . . . . . . . . . . . . . . . . . . . . . .
446
www.bjsportmed.com
Downloaded from bjsm.bmjjournals.com on 22 October 2006
METHODS
Data collection
Player injuries were prospectively recorded over the period
November 1994 to May 1997 at four English professional football
clubs.All professional players at these clubs were included
in the study together with youth players from two of the clubs.
Each injury was diagnosed by the clubs’ senior physiotherapists,
who were qualified to either Chartered status or FA
Diploma level. The senior physiotherapist at each of the four
clubs recorded their injury data on a specific injury report
form designed for this study.4 A recordable injury was defined
as an injury received during competition or training that prevented
the player from participating in competition or normal
training for at least one day, not including the day of the
injury. These injuries were categorised as “slight” (one to three
days), “minor” (four to seven days), “moderate” (one to four
weeks), and “major” (more than four weeks). The categories
designated as “minor”, “moderate”, and “major” corresponded
to work based injuries that would be defined as
reportable injuries under the Reporting of Injuries, Diseases,
and Dangerous Occurrences Regulations.13 For every injury,
the physiotherapists were also required to record the nature,
location, and mechanism of the injury and the specific
number of days that players were unavailable to train or to
play as a result of each injury. Absences caused by sickness or
other general medical conditions,which required an examination
by the player’s own general practitioner, were not
included in the study.
Comparative data for work based fatalities and major and
over three day injuries for the construction, manufacturing,
and service industries were obtained from statistics published
by the Health and Safety Commission.3
Data analysis
Probability is defined as the frequency of occurrence of an
adverse event and, in the present context, was expressed as
the number of events per 100 000 hours of exposure. The six
levels of probability used in this study were set over the range
from 0.001 to 1000 events per 100 000 hours of exposure in
order to correspond to the probability values observed in football
and a range of other occupations (fig 1). The consequences
of an injury were defined principally in terms of the number of
days of absence from training and/or competition experienced
by the player, because this reflected the impact of injury on
both players and clubs. Data for injury location, nature, and
mechanism were analysed and reported as proportions of the
total number of injuries and of the total risk and as percentage
cumulative frequencies of the number of injuries and of the
risk from injuries. The risk (Rd) associated with a specified
number of days absence (d) was defined as the total number
of injuries (nd) resulting in the specified number of days of
absence multiplied by the specified number of days:
Figure 1 Injury risk matrix.
Superscripts are reference numbers.
OA, Osteoarthritis.
A risk based assessment of football injuries 447
www.bjsportmed.com
Downloaded from bjsm.bmjjournals.com on 22 October 2006
Rd = d ´ nd
The total risk (RT) within the category of injury being assessed
was defined by the sum of Rd for all values of d from d = 1 to
¥:
RT = SRd = S(dnd)
To compare risk levels in football with other occupational sectors,
the RIDDOR reportable category of over three days
absence from work was taken to equate to the categories of
minor and moderate injuries reported here, and the RIDDOR
reportable category of major injury was equated to the major
injury category reported here.13
The consequences of injury were also evaluated against an
estimated equivalent monetary value and, in this respect, the
value of a life (fatality) was taken to range from £1m to £10m,
which is a typical range of values used by the UK government
and industry for the purposes of cost benefit analysis. The
monetary values used for the less severe categories of major,
moderate, minor, and slight injuries were set at decreasing
exponential intervals down to the range of £100 to £1000 for a
slight injury resulting in one to three days of absence. The
values for days lost also equate to the equivalent monetary
values of the days lost based on typical salaries of professional
footballers.14
The boundaries of the evaluation areas within the risk
matrix were defined by the HSE criteria for the acceptability
and tolerability of risk in terms of the probability of a fatal
accident,12 with the boundaries at lower levels of consequence
defined by equivalent levels of risk based on the product of
probability and financial losses.
Statistical analysis
A one way c2 test was used to assess differences between the
results obtained for the proportions of injuries and proportions
of risk. Significance was accepted at the 95% confidence
level for all statistical tests, unless stated otherwise.
RESULTS
Overview
The injury database covered 138 players and included 744
reports of injury that prevented a player from training or competing
for at least one day; a detailed analysis of the
epidemiological data, in terms of injury nature, location, and
mechanism, has been reported separately.4 Of the 744 injuries
reported, 587 (76%) were accompanied by information on the
number of days that the player missed from training and
competition as a result of the injury. These injuries accounted
for a total of 8644 days of absence, which equated to a mean
(SD) absence period of 14.7 (22.5) days per injury (new injuries,
13.7 days; re-injuries, 18.2 days) and were equivalent to
39.6 days of absence per player per year or 13% of a playing
season. Re-injuries accounted for 22.3% of all the injuries
recorded.
Injury frequency and severity
Table 1 summarises the proportions and rates of injury and the
proportions of time lost, as a function of injury severity. There
were significant differences in the proportions of time lost
through injury as a function of injury severity (p<0.001). Figure
2 shows the frequency distributions for the number of
days lost (up to eight weeks) for all new and re-injuries, and
figure 3 shows the percentage cumulative frequency and percentage
cumulative risk for new injuries as a function of days
lost.
Table 1 Overview of the risks of injury as a function
of injury severity
Injury severity
Proportion of
injuries (%)
Proportion of
time lost through
injury (%)
Injury rate per
100000 hours
Slight 14.1 3.3 140
Minor 35.4 14.3 300
Moderate 38.0 35.8 320
Major 12.4 46.6 90
Total 100.0 100.0 850
Figure 2 Frequency distributions for the days of absence as the result of new injuries and re-injuries.
Figure 3 Percentage cumulative frequency and cumulative risk of
injury as a function of days of absence.
448 Drawer, Fuller
www.bjsportmed.com
Downloaded from bjsm.bmjjournals.com on 22 October 2006
Table 2 gives the proportions of injuries and accident rates
for fatalities and major and over three day injuries from the
construction, manufacturing, and service industries, which
were calculated from data published by the Health and Safety
Commission.3
Injury location, nature, and mechanism
Tables 3, 4, and 5 present the percentages of injuries and levels
of risk within the five main categories of injury location
(thigh, ankle, knee, lower leg, and groin), the four main
categories of injury nature (strains, contusions, sprains, and
fractures/dislocations), and the four main categories of injury
mechanism (tackled, running, tackling, and shooting) respectively.
Player to player contact incidents, such as tackling and
being tackled, accounted for 38.9% of the total days of
absence, and football specific activities, such as running and
shooting,were responsible for 46.5% of the days lost. Player to
player contact was responsible for 93.8% of contusions, 59.4%
of sprains, and 55.0% of fracture injuries. The players making
the tackles suffered all of the fracture injuries caused through
player to player contact. Player to player contact was also the
major cause of ankle (61.9%) and knee (56.8%) injuries, and
football specific activities were the major causes of strains
(72.1%) and groin (46.8%), thigh (71.3%), and lower leg
(43.6%) injuries.
Percentage cumulative frequency and cumulative risk
distributions
Table 6 shows the numbers of days of absence at which 50% of
the injuries and 50% of the risk occurred for new and
re-injuries and for injury location, nature and mechanism.
Table 6 also shows the percentages of injuries at which 50% of
the risk occurred within each of these categories.
Table 2 Proportions of injuries and injury rates in the construction, manufacturing,
and service sectors as a function of injury severity
Industrial sector Injury severity Proportion of injuries (%)
Injury rate per 100000
hours
Construction Over 3 days 69.6 0.45
Major 30.0 0.20
Fatal 0.41 0.003
All 100.0 0.65
Manufacturing Over 3 days 83.0 0.49
Major 16.9 0.10
Fatal 0.08 0.0005
All 100.0 0.59
Service Over 3 days 84.2 0.21
Major 15.7 0.039
Fatal 0.05 0.0001
All 100.0 0.25
Table 3 Distribution of injuries and risk as a function of injury location
Injury location
Thigh Ankle Knee Lower leg Groin
Proportion of injuries (%) 22.2 16.0 15.2 13.0 10.8
Proportion of risk (%) 18.5 15.1 22.2 13.4 10.9
Table 4 Distribution of injuries and risk as a function of injury nature
Injury nature
Strains Contusions Sprains
Fractures/
dislocations
Proportion of injuries (%) 40.6 19.8 19.3 3.8
Proportion of risk (%) 36.2 10.1 25.7 9.4
Table 5 Distribution of injuries and risk as a function of injury mechanism
Injury mechanism
Tackled Running Tackling Shooting
Proportion of injuries (%) 23.5 19.6 12.2 10.3
Proportion of risk (%) 23.7 17.4 12.7 11.9
A risk based assessment of football injuries 449
www.bjsportmed.com
Downloaded from bjsm.bmjjournals.com on 22 October 2006
DISCUSSION
Frequency based assessments of injury only describe the likelihood
or probability that an injury will occur, whereas risk is
a function that takes account of both probability and the consequences
resulting from an adverse event. Although the
probability of an adverse event occurring will be the same for
all stakeholders in a football club, quite often each stakeholder
will be concerned about and affected by different outcomes
from a player’s injury. For example, a knee injury may lead to
osteoarthritis in the knee of the player, a heavier treatment
and rehabilitation workload for the physiotherapist, a poorer
performance by the team, and a worse financial performance
by the club. Therefore, for an accurate evaluation of the impact
of players’ injuries, measures of both injury severity and probability
should be included within an assessment. In this
respect, injury severity or consequence can be defined in a
number of ways, such as the nature of the injury, the duration
of the injury treatment and rehabilitation processes, and the
financial loss experienced by the club and/or player.
If a simple comparison is made between the frequency based
and risk based distributions of injuries shown in tables 3, 4, and
5, there appears at first to be little difference between the results
obtained from the two approaches in the cases of injury location
and mechanism.For injury nature, the risk from contusionswas
half the level indicated by the frequency of occurrence,whereas
the risk from fractures/dislocations was three times higher than
the level indicated by the frequency of occurrence. The results
presented in tables 1 and 6, however, emphasise the importance
of taking risk based as well as frequency based assessments of
injury into account. Table 1, for example, shows that if the
severities of players’ injuries are ranked in decreasing order of
frequency of occurrence, they appear as moderate, minor, slight,
and major injuries, whereas, if the severities are ranked in
decreasing order of risk they appear as major,moderate, minor,
and slight injuries. In addition, whereas slight and minor injuries
contributed 51% of the number of injuries, they represented
only 17% of the total risk, whereas major injuries only
accounted for 12% of the injuries but 47% of the risk. Similarly,
the results presented in table 6 show that, apart from fractures/
dislocations, 50% of the injuries in each category resulted in 12
or less days of absence. However, injuries included within the
50% risk value in each category lasted for up to 42 days. In addition,
between 8% (lower leg) and 19% (ankle) of injuries,which
represented the more severe injuries, accounted for 50% of the
total risk. As major injuries normally require very different
treatment and rehabilitation programmes from slight and
minor injuries, these results are particularly important as far as
decisions on the allocation of resources for injury prevention,
treatment, and rehabilitation support services are concerned.
An assessment of the levels of support services provided by
professional football clubs in the United Kingdom showed
that, although the provision and application of injury
treatment and rehabilitation services were adequate, the provision
of injury prevention services was inadequate.10 Although
deficiencies in the numbers, qualifications, and
experience of medical staff at football clubs have been
reported as an important issue,15 the underlying cause of the
problems associated with the provision of injury prevention
can often be traced to the financial pressures experienced by
most professional clubs.14 This situation has been exacerbated
by the short term management strategy adopted by many
clubs whereby the limited financial resources that are
available are prioritised and allocated in favour of the recruitment
and remuneration of players,14 which is supported by a
strategy of treatment and rehabilitation for players when they
are injured.10 The longer term benefits offered by a risk management
approach,which advocates the allocation of adequate
resources to injury prevention strategies, have therefore not
gained currency among most English clubs.
An effective risk management strategy begins with an
estimation and evaluation of the risks associated with the activity.
However, to complete this assessment, it is necessary to use
acceptable criteria for the estimation and evaluation of the risk.
Risk, which is described as a function of probability and consequence,
can be displayed on a two dimensional calibrated
matrix, such as the 5 (consequence) ´ 6 (probability) matrix
used here. The data for acute football injuries (table 1) and the
construction, manufacturing, and service sectors of industry
(table 2) are presented within the 5 ´ 6 risk matrix shown in fig
1. In addition, risk data have been included for the occurrence of
osteoarthritis in the lower limb joints of retired footballers by
taking the probability of occurrence5 (two cases/100 000 hours)
and assuming that the consequences of osteoarthritis are
equivalent to a major acute injury. The data for fatalities in the
manufacturing and service sectors fell within the tolerable
range of risk, but the probability values were below the range
included in the risk matrix presented here. Fatalities in the construction
industry and major injuries in the construction,
manufacturing, and service sectors fell within the area defined
by the HSE as tolerable, and the data for over three day injuries
in the construction, manufacturing, and service sectors fell
within the acceptable and tolerable areas of risk. For football
Table 6 Fifty per cent cumulative frequency and risk values as a function of
causative factors
Causative factor
Number of days of
injury at 50%
cumulative frequency
Number of days of
injury at 50%
cumulative risk
Percentage of
injuries at the 50%
cumulative risk level
New injuries 8 28 86
Re-injuries 10 44 91
Injury location
Groin 10 30 85
Thigh 8 18 86
Knee 12 42 87
Lower leg 6 42 92
Ankle 10 20 81
Injury nature
Strains 8 20 86
Sprains 10 32 86
Contusions 6 12 83
Fractures/ dislocations 40 90 81
Injury mechanism
Player to player contact 8 28 88
Football related 8 22 86
450 Drawer, Fuller
www.bjsportmed.com
Downloaded from bjsm.bmjjournals.com on 22 October 2006
injuries, however, although the occurrence of slight injuries fell
within the tolerable level of risk, the occurrence of minor,moderate,
and major acute injuries and osteoarthritis in the lower
limb joints all fell within the area of risk deemed to be
unacceptable by the HSE.
These unacceptable levels of risk to players from acute and
chronic injuries should be amajor concern for all stakeholders
in English professional football. Table 1 shows that nearly 50%
of the total risk from injuries was associated with the 12% of
major injuries; therefore significant reductions in the overall
level of risk could be achieved if the treatment and rehabilitation
programmes for these long terminjuries could be reduced
significantly. However, 22% of all injuries were attributable to
re-injuries, which indicated that caution should be exercised
before introduction of non-validated treatment and rehabilitation
programmes. Inadequate rehabilitation and incomplete
healing of injuries have been suggested as key reasons for the
high levels of re-injury in football.16 In general, ankle and knee
sprains and thigh strains have been identified as the injuries
with the greatest chance of re-injury.17 Re-injuries are reported
to be more severe than the original injury,18 and the results
obtained from this study support this, as the average period of
absence for re-injuries was 33% greater than that observed for
new injuries. The probability of a re-injury occurring could be
reduced if the sports science and medical team measured each
player’s levels of fitness and physical condition when they
were fully fit and used these values as benchmarks to assess a
player’s level of recovery during an injury treatment and rehabilitation
programme.
The high proportion of risk associated with player to player
contact indicated that players had a high level of accountability
for their own and other players’ injuries. More effective
coaching, which would improve tackling skills, and appropriate
education for players about the personal and third party consequences
of poor tackling technique could both contribute to
significant reductions in the number and severity of contact
injuries. In addition, the number of player to player contact
injuries could be further reduced through more effective control
of players by referees and through the use of more stringent
penalties for players causing injuries as a result of foul play. The
proportion and level of risk associated with football specific
activities indicated that many players were not adequately prepared
for training and competition. Therefore, coaches and
fitness trainers should take into account current sports science
knowledgewhen developing balanced fitness and skills training
programmes. However, it is the chairmen who define the allocation
of financial resources to injury prevention, and the team
managers who define the team’s playing culture within a football
club, and as such they have the greatest influence on and
responsibility for the levels of injury in professional football. It is
essential that a club’s financial resources are allocated and utilised
efficiently and effectively in order to reduce the overall risk
to clubs and players.Drawer and Fuller19 have developed an economic
framework, based on the relations between the availability
and quality of players, team performance, club turnover, and
team salaries, that can be applied to cost benefit analyses.However,
by taking into account the number of players at a club, the
rates of injury, and the equivalent monetary values at each level
of injury severity, it is possible to obtain a relatively simple
assessment of the costs associated with injury, which can then
be used for evaluating and optimising resource allocations
within a range of injury control strategies.
Conclusions
The results from this study have highlighted that the HSE
would deem the levels of risk associated with professional football
to be unacceptable. The high proportion of risk attributable
to re-injuries illustrates the potential impact thatmanagersmay
have when selecting players for competition before they have
fully recovered from injury. The sports science and medical team
has a major role to play in minimising the risks associated with
re-injuries by developing effective and efficient injury rehabilitation
programmes. The players, however, have a significant
responsibility for the level of injury because of their poor standards
of tackling and preparation for training and competition.
Refereesmust also accept responsibility for their contribution to
the level of injury through their role as arbitrators of the laws of
the game, which have been developed to protect players from
injury. Finally, the sport’s administrators must ensure that suspension
and financial punishments are used to provide an
adequate deterrent to those players whose poor technique or
inappropriate tackling results in serious injuries to other
players.
. . . . . . . . . . . . . . . . . . . . .
Authors’ affiliations
S Drawer, UK Sports Institute, 40 Bernard Street, London WC1N 1ST,
UK
C W Fuller, Scarman Centre, University of Leicester, Leicester LE1 7QA,
UK
REFERENCES
1 HMSO. Health and safety at work etc Act. London: HMSO, 1974.
2 HMSO. Management of health and safety at work regulations. London:
HMSO, 1999.
3 Health and Safety Commission. Health and safety statistics
1999/2000. London: HMSO, 2000.
4 Hawkins RD, Fuller CW. A prospective epidemiological study of injuries
in four English professional football clubs. Br J Sports Med
1999;33:196–203.
5 Drawer S, Fuller CW. An investigation into the propensity for
osteoarthritis and lower limb joint-pain in retired professional soccer
players. Br J Sports Med 2001;35:402–8.
6 Ekstrand J, Gillquist J. The avoidability of soccer injuries. Int J Sports
Med 1983;4:124–8.
7 Engstrom B, Johansson C, Tornkvist H. Soccer injuries among elite
female players. Am J Sports Med 1991;19:372–5.
8 Hawkins RD, Fuller CW. An examination of the frequency and severity
of injuries and incidents at three levels of professional football. Br J Sports
Med 1998;32:326–32.
9 Hawkins RD, Fuller CW. A preliminary assessment of professional
footballers’ awareness of injury prevention strategies. Br J Sports Med
1998;32:140–3.
10 Drawer S, Fuller CW. Benchmarking the levels of injury support services
available at English professional football clubs. European Journal for
Sport Management 1999;6:34–47.
11 Health and Safety Executive. Generic terms and concepts in the
assessment and regulation of industrial risks. Sudbury: HSE Books, 1995.
12 Health and Safety Executive. The tolerability of risks from nuclear
power stations. Sudbury: HSE Books, 1988.
13 HMSO. Reporting of injuries, diseases and dangerous occurrences
regulations 1995. London: HMSO, 1995.
14 Boon G, Eaves N, Jones D, et al. England’s premier clubs. Manchester:
Deloitte & Touche, 2001.
15 Waddington I, Roderick M, Naik R. Methods of appointment of club
doctors and physiotherapists in English professional football clubs: some
problems and issues. Br J Sports Med 2001;35:48–53.
16 Inklaar H. Soccer injuries. II. Aetiology and prevention. Sports Med
1994;18:81–93.
17 Inklaar H. Soccer injuries. I. Incidence and severity. Sports Med
1994;18:55–73.
18 Hawkins RD, Hulse MA, Wilkinson C, et al. The association football
medical research programme: an audit of injuries in professional football.
Br J Sports Med 2001;35:43–7.
19 Drawer S, Fuller CW. An economic framework for assessing the impact
of injuries in professional football. Safety Science 2001;39:in press.
Take home message
The risk of injury to professional footballers has been
assessed against the risk criteria of negligible, acceptable,
tolerable, and unacceptable, which are used by regulators
when assessing risks in other industrial and commercial
sectors. The levels of risk associated with acute and
chronic injuries are shown to fall within the unacceptable
level of risk.
A risk based assessment of football injuries 451
www.bjsportmed.com
Downloaded from bjsm.bmjjournals.com on 22 October 2006

original article
The
new england journal
of
medicine
n engl j med
352;5
www.nejm.org february
3, 2005
468
A Clone of Methicillin-Resistant
Staphylococcus aureus
among Professional
Football Players
Sophia V. Kazakova, M.D., M.P.H., Ph.D., Jeffrey C. Hageman, M.H.S.,
Matthew Matava, M.D., Arjun Srinivasan, M.D., Larry Phelan, B.S., B.A.,
Bernard Garfinkel, M.D., Thomas Boo, M.D., Sigrid McAllister, B.S., M.T.(A.S.C.P.),
Jim Anderson, B.S., A.T.C., Bette Jensen, M.M.Sc., Doug Dodson, B.S.,
David Lonsway, M.M.Sc., Linda K. McDougal, M.S., Matthew Arduino, Dr.P.H.,
Victoria J. Fraser, M.D., George Killgore, Dr.P.H., Fred C. Tenover, Ph.D.,
Sara Cody, M.D., and Daniel B. Jernigan, M.D., M.P.H.
From the Division of Healthcare Quality
Promotion, National Center for Infectious
Diseases (S.V.K., J.C.H., A.S., S.M., B.J., D.L.,
L.K.M., M.A., G.K., F.C.T., D.B.J.), and the
Epidemic Intelligence Service, Division of
Applied Public Health Training, Epidemiology
Program Office (S.V.K., T.B.), Centers
for Disease Control and Prevention, Atlanta;
the Departments of Orthopedic Surgery
(M.M.) and Internal Medicine (B.G.) and
the Department of Internal Medicine, Infectious
Diseases Division (V.J.F.), Washington
University School of Medicine, St.
Louis; the Missouri Department of Health
and Senior Services, St. Louis (L.P., D.D.);
the BJC Medical Group, St. Louis (B.G.); the
St. Louis Rams Professional Football Team,
St. Louis (J.A.); and the Office of Disease
Control, Santa Clara County Health Department,
San Jose, Calif. (S.C.). Address reprint
requests to Dr. Kazakova at the Epidemic
Intelligence Service, Division of Healthcare
Quality Promotion, National Center for Infectious
Diseases, CDC, 1600 Clifton Rd,
MS A35, Atlanta, GA 30333, or at srk7@
cdc.gov.
N Engl J Med 2005;352:468-75.
Copyright © 2005 Massachusetts Medical Society.
background
Methicillin-resistant
Staphylococcus aureus
(MRSA) is an emerging cause of infections
outside of health care settings. We investigated an outbreak of abscesses due to MRSA
among members of a professional football team and examined the transmission and
microbiologic characteristics of the outbreak strain.
methods
We conducted a retrospective cohort study and nasal-swab survey of 84 St. Louis Rams
football players and staff members.
S. aureus
recovered from wound, nasal, and environmental
cultures was analyzed by means of pulsed-field gel electrophoresis (PFGE) and
typing for resistance and toxin genes. MRSA from the team was compared with other
community isolates and hospital isolates.
results
During the 2003 football season, eight MRSA infections occurred among 5 of the 58
Rams players (9 percent); all of the infections developed at turf-abrasion sites. MRSA
infection was significantly associated with the lineman or linebacker position and a
higher body-mass index. No MRSA was found in nasal or environmental samples; however,
methicillin-susceptible
S. aureus
was recovered from whirlpools and taping gel
and from 35 of the 84 nasal swabs from players and staff members (42 percent). MRSA
from a competing football team and from other community clusters and sporadic cases
had PFGE patterns that were indistinguishable from those of the Rams’ MRSA; all carried
the gene for Panton–Valentine leukocidin and the gene complex for staphylococcal-
cassette-chromosome
mec
type IVa resistance (clone USA300-0114).
conclusions
We describe a highly conserved, community-associated MRSA clone that caused abscesses
among professional football players and that was indistinguishable from isolates
from various other regions of the United States.
abstract
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3, 2005
methicillin-resistant
s. aureus
among football players
469
ontact sports such as american
football inevitably lead to skin and soft-tissue
injuries that place players at increased
risk for infection.
1,2
Skin infections, particularly
those caused by
Staphylococcus aureus
, are common
among sports participants. Recent reports have described
an increasing number of community-associated
methicillin-resistant
S. aureus
(MRSA) skin
infections in persons without links to health care institutions.
3-6
These infections differ from those due
to health care–associated MRSA in that they are
resistant predominantly to
b
-lactam and macrolide
antimicrobial agents and in that they result in
the formation of skin abscesses. Cases of community-
associated MRSA infection have previously been
reported among football players and other sports
participants; however, little is known about factors
associated with the emergence of communityassociated
MRSA strains that may cause outbreaks
in various geographic regions and community settings.
In September 2003, cases of large skin abscesses
caused by MRSA were first noted among members
of the St. Louis Rams, a professional football team
in Missouri. Additional cases among team members
and subsequent cases in members of an opposing
team suggested that competitive play might
be causing transmission. On November 6, 2003, the
Centers for Disease Control and Prevention (CDC)
was invited to investigate the transmission of MRSA
among the Rams football players, to recommend
prevention and control measures, and to characterize
the staphylococcal isolates.
epidemiologic investigation
We defined a case of MRSA infection as any skin
or soft-tissue infection in a player or staff member
of the St. Louis Rams during the 2003 football season
(August 1 through November 30) from which
MRSA was isolated on culture. To identify potential
activities that might have led to the transmission of
MRSA, we performed an observational study of onfield
and off-field activities and hygiene practices
during competition and training at the Rams facility.
In addition, a retrospective cohort study of
the players was conducted to identify risk factors
for infection. Using a standardized data-collection
form, we collected information about players’ field
positions, demographic characteristics, health care
exposures, antimicrobial use, close contact with
other persons with skin infections, skin-abrasion
management, hygiene practices, and use of saunas,
whirlpools, and training and therapy equipment.
We also evaluated antimicrobial use among the
Rams players by reviewing the team pharmacy log
and calculating the average number of antimicrobial
prescriptions per player per year. We compared
this rate to sex- and age-specific rates in the general
population, as determined by national surveys.
7,8
environmental and laboratory
investigation
To determine whether other Rams players were
colonized with the outbreak MRSA strain, we performed
a nasal-carriage prevalence survey among
all players and staff members and obtained swabs
of uninfected skin abrasions. To identify any environmental
sources of exposure, we sampled surfaces
and shared items in the training facility, including
weight-training equipment, towels, saunas
and steam rooms, and water from whirlpools and
a therapy pool. In addition, we swabbed 0.1-m
2
(1-ft
2
) areas of artificial turf after a game in areas of
the field that were recorded to have the highest number
of tackles. All environmental sampling was performed
after recommended infection-control procedures
and the use of chlorhexidine-containing
soap had been initiated.
After initial screening for oxacillin resistance, all
available MRSA isolates from Rams players’ skin
abscesses and suspected MRSA isolates from skin
abrasions, nasal swabs, and environmental sources
were tested for antimicrobial susceptibility by
means of broth microdilution, according to interpretive
criteria of the Clinical and Laboratory Standards
Institute (formerly the National Committee
for Clinical Laboratory Standards).
9
The antimicrobial
agents tested were ciprofloxacin, clindamycin,
erythromycin, levofloxacin, oxacillin, penicillin,
tetracycline, trimethoprim–sulfamethoxazole, and
vancomycin. In addition, an antimicrobial-susceptibility
disk test to study inducible clindamycin resistance
(i.e., a D-test) was performed.
10,11
All
S. aureus
recovered from skin abscesses, the environment,
nasal swabs, and skin-abrasion specimens
were tested by a polymerase-chain-reaction
assay for the staphylococcal-cassette-chromosome
mec
(SCC
mec
) resistance complex, as described by
Katayama et al.,
12
and for the gene encoding Panton–
Valentine leukocidin cytotoxin.
13,14
After digestion of chromosomal DNA with restriction
endonucleases (
Sma
I for initial digestion
c
methods
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3
,
2005
The
new england journal
of
medicine
470
and
Eag
I,
Sac
II,
Nar
I,
Apa
I, or
Nae
I for subsequent
digestion of subtype USA300-0114), restriction
products were analyzed by pulsed-field gel electrophoresis
(PFGE).
15
The gels were analyzed with
BioNumerics software (Applied Maths) and interpreted
according to criteria published elsewhere.
16
To determine the relatedness of the outbreak strain
to other strains, we compared the PFGE patterns of
isolates from the Rams with those of MRSA isolates
from Team A, a professional football team that
competed with the Rams and members of which
subsequently had abscess development. In addition,
we compared the PFGE patterns of both teams’ isolates
with all 3241 isolates of
S. aureus
in the CDC
PulseNet database
17
to identify the clonal complex
to which the isolates belonged. To compare the
patterns, we calculated percentage similarities with
Dice coefficients by the unweighted pair-group
method with arithmetic averages.
17
Multilocus sequence
typing was performed on pulsed-field type
USA300 isolates.
18
statistical analysis
All univariate and bivariate analyses were performed
with SAS software, version 9.0. Chi-square
or Fisher’s exact tests were used to analyze the relationships
between categorical variables, and t-tests
were used to analyze the relationship between categorical
and continuous variables. All reported P values
are two-sided. Multivariate analysis was not performed
because of the small number of cases.
epidemiologic investigation
From September 1 through December 1, 2003,
eight MRSA infections occurred in 5 of the 58 Rams
players (9 percent) (Fig. 1). The infections developed
in offensive and defensive linemen and a linebacker
at sites of skin abrasions (turf burns) on elbows,
forearms, or knees. All the infections rapidly
progressed to large abscesses 5 to 7 cm in diameter
and required surgical intervention with incision
and drainage. The mean age of the players with
MRSA infections was 27 years (range, 23 to 33). Various
antimicrobial agents were administered; two
of the players received intravenous antimicrobial
agents (vancomycin and ceftriaxone) before the initiation
of oral antimicrobial therapy, and all five players
received three oral agents (cephalexin, trimethoprim–
sulfamethoxazole, and rifampin) alone or
in combination. Most of the infections resolved
within 10 days after the initiation of treatment.
Recurrent infections developed in three of the five
players. Although none of the players required hospitalization,
three of them missed 1, 4, and 12 days
of games or practice, respectively, for a total of 17
missed days due to infection.
From our player survey and observational study
of games and practices, we found that skin abrasions
occurred frequently among players. Approximately
two to three turf burns per week were acquired
from sliding on the field during competition
or practice (Fig. 2). Players reported that abrasions
were more frequent and severe when competition
took place on artificial turf than when it took place
on natural grass. Trainers, who provided wound
care, did not have regular access to hand hygiene,
and alcohol-based hand-hygiene products were
not available near areas where wound care or physical
therapy was provided. Towels were frequently
shared on the field during practice and games,
with as many as three players using the same towel.
Players often did not shower before using communal
whirlpools. At the training facility, weighttraining
and therapy equipment was not routinely
cleaned. Manufacturer-recommended guidelines
for the routine cleaning of whirlpools, saunas, and
steam rooms were not available for review.
Evaluation of potential risk factors explored in
a cohort study revealed that being a lineman or a
linebacker, as compared with having a backfield
position, was associated with the highest relative
risk of an MRSA infection (10.6 [95 percent confidence
interval, 1.3 to infinity], P=0.02) (Table 1).
Players with MRSA skin infection had a significantly
higher body-mass index than players in whom infection
did not develop. Use of antimicrobial agents
during the previous year was associated with MRSA
infection; however, the association was not statistically
significant (relative risk, 7.8; 95 percent confidence
interval, 0.5 to infinity).
According to the team pharmacy log for the
2002 football season, maintained at the training
facility, a team player on average received 2.6 antimicrobial-
drug prescriptions per year. This rate was
greater than 10 times the rate among persons of
the same age and sex in the general population (0.5
prescription per year). In their survey responses,
approximately 60 percent of players indicated they
had taken or received antimicrobials during the
2003 football season.
Infection-control measures were instituted at
the Rams training facility during the week of Octoresults
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3, 2005
methicillin-resistant
s. aureus
among football players
471
ber 26, 2003 (Fig. 1), and included installation of
wall-mounted soap dispensers with chlorhexidinecontaining
soap for routine hand washing by players
and staff members, appropriate local wound
care, antimicrobial therapy targeting MRSA, and
active surveillance for skin infections. After this intervention,
only one additional case of MRSA infection
occurred.
laboratory investigation
Susceptibility testing of the MRSA causing infections
in five Rams players showed that, in all cases,
it was resistant to macrolides and oxacillin but susceptible
to ciprofloxacin, clindamycin, tetracycline,
trimethoprim–sulfamethoxazole, and vancomycin.
None of the tested isolates exhibited inducible
clindamycin resistance on the D-test. Isolates
from the Rams and from Team A were compared
with 3241
S. aureus
isolate patterns in the CDC staphylococcus
PulseNet database, which revealed that
both teams’ isolates were pulsed-field type USA300
(Fig. 3). After digestion with
Sma
I endonuclease,
the teams’ PFGE patterns were indistinguishable
from one another but also indistinguishable from
patterns associated with various community-associated
MRSA clusters and sporadic cases in the United
States. Both the teams’ isolates and the indistinguishable
community-associated isolates differed
from other community-associated MRSA isolates
(USA400) and known health care–associated isolates
(USA100 and USA200).
To discriminate further among the indistinguishable
PFGE patterns, digestion with five additional
enzymes (
Eag
I,
Sac
II,
Nar
I,
Apa
I, and
Nae
I) was
performed on MRSA isolates: two from the Rams,
eight from Team A, and one representative isolate
from each of nine previously investigated community
clusters and sporadic cases among sports par-
Figure 1. Epidemic-Curve Graph (Top) and Field Position Diagram (Bottom) of Cases of MRSA Infection among St. Louis
Rams Professional Football Players in 2003.
Each box on the epidemic-curve graph and field diagram represents an MRSA infection; different colors designate different
players; boxes of the same color thus represent recurrent infections. On the field diagram, X represents a defensiveplayer
position and O an offensive-player position.
Offense
Defense
No. of Infectionss
2
1
0
3 10 17 24 31 7 14 21 28 5 12 19 26 2 9 16 23 30
August September October
Week of Infection
November
First MRSA
infection
in Team A
Game
against
Team A
Initiation of infectioncontrol
measures
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3
,
2005
The
new england journal
of
medicine
472
ticipants, children, prisoners, military recruits, and
men who have sex with men. After each of the five
additional digestions, the isolates again had indistinguishable
patterns. This clonal subtype is now
classified as pulsed-field type USA300-0114. All
USA300-0114 isolates contained the gene for Panton–
Valentine leukocidin as well as the gene complex
for SCC
mec
type IVa resistance. The USA300-
0114 subtype has been determined to fall within
sequence type 8 on multilocus sequence typing.
The nasal-swab survey indicated that 23 of the
58 Rams players (40 percent) and 12 of the 26 staff
members (46 percent) were colonized with methicillin-
susceptible
S. aureus
(MSSA). No MRSA was
identified. No environmental specimens yielded
MRSA; however, MSSA isolates were recovered from
a gel-applicator stick used for taping ankles and
from two samples of whirlpool water collected at
the end of the day. The gel and whirlpool-water isolates
were indistinguishable on PFGE from MSSA
recovered from the nasal swabs.
Examination of one nasal MSSA isolate revealed
a two-band difference from the USA300-0114 pattern.
This isolate also had the gene for Panton–
Valentine leukocidin. Southern blot hybridization
analysis of the outbreak MRSA isolates and the nasal
MSSA isolate demonstrated that the band missing
in the MSSA isolate carried the
mecA
resistance
gene. The absence of the
mecA
gene may indicate either
that the isolate lost the resistance gene or that
this MSSA represents a strain that has not yet acquired
the resistance gene.
Our investigation revealed that a cluster of skin abscesses
among professional football players and
other recent outbreaks of skin infection in the United
States were caused by an emerging MRSA clone.
This community-associated clone differed from
strains of MRSA circulating in health care settings
in that it was susceptible to most antimicrobial
agents other than
b
-lactams and macrolides, it primarily
caused skin infections in otherwise healthy
persons, and it carried both a characteristic gene
complex for methicillin resistance (SCC
mec
type
IVa) and the gene for Panton–Valentine leukocidin,
a cytotoxin that has been associated with severe abscesses
and necrotizing pneumonia.
13
During the
2003 football season, abscesses also occurred in a
competing team after a game with the Rams, suggesting
that transmission of MRSA occurred during
game play.
We used PFGE and five restriction endonucleases
to demonstrate that MRSA isolates from both
teams were indistinguishable. Further comparison
with MRSA from epidemiologically unrelated outbreaks
among sports participants and persons in
other settings revealed that these other isolates were
also indistinguishable from those of the two professional
football teams and represented a clone
now classified as pulsed-field type USA300-0114.
These results indicate that this clone may be widely
distributed in the community and thus that the two
teams may have acquired the same strain independently.
With currently available molecular-typing
methods, it was not possible to differentiate between
community and team isolates, and thus neither
team-to-team transmission nor independent,
community acquisition could be implicated as the
primary source of MRSA among football players.
discussion
Figure 2. Photograph of an Uninfected Skin Abrasion
(Turf Burn) on a St. Louis Rams Professional Football
Player in 2003.
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3, 2005
methicillin-resistant
s. aureus
among football players
473
Findings from our investigation underscore the
importance of certain factors at the player level
and at the team level that could have facilitated the
spread of the clone in this setting. One important
player-level factor was skin abrasions, or turf burns.
All MRSA skin abscesses developed at sites of turf
burns on areas of skin not covered by a uniform
(e.g., elbows and forearms). These abrasions were
usually left uncovered, and when combined with
frequent skin-to-skin contact throughout the football
season, probably constituted both the source
and the vehicle for transmission. In our investigation,
infection occurred only among linemen and
linebackers, and not among those in backfield positions,
probably because of the frequent contact
among linemen during practice and games. We also
observed a lack of regular access to hand hygiene
(i.e., soap and water or alcohol-based hand gels)
for trainers who provided wound care; skipping of
showers by players before the use of communal
whirlpools; and sharing of towels — all factors
that might facilitate the transmission of infection
in this setting.
We did not detect any MRSA in environmental
or nasal samples; however, environmental sources
yielded MSSA that matched nasal MSSA isolates
— a finding that suggests that the environment
may have had a role in the transmission of MRSA
among team members. Previously reported investigations
have identified potential transmission
from contaminated surfaces and shared items.
23
In addition, recovery of MRSA from colonized persons
during outbreaks has been variable, with some
investigations detecting no nasal colonization with
MRSA,
19,22
as was the case in our study. Treatment
of infected players with antimicrobial drugs (e.g.,
rifampin) in our study may have eliminated the nasal
colonization in these persons. In addition, institution
of infection-control practices and enhancement
of personal hygiene may have minimized
* The data reflect information provided by the players who responded to the survey. CI denotes confidence interval, and NA not
applicable. Percentages and relative risks were calculated on the basis of the total number of responses to each question.
† P values and confidence intervals are based on Fisher’s exact and chi-square analysis for categorical variables and t-testing
for continuous variables.
‡ The mean body-mass index is the weight in kilograms divided by the square of the height in meters.
Table 1. Risk Factors for Skin Abscesses Due to Community-Associated MRSA among 53 St. Louis Rams Football Players,
August 1 through November 30, 2003.*
Risk Factor Risk Factor Present
Relative Risk
(95% CI) P Value†
All
Respondents
MRSA
Infection
No MRSA
Infection
Black race — no. of players (%) 25/53 (47) 3/5 (60) 22/48 (46) 1.7 (0.3–9.3) 0.55
Mean body-mass index‡ NA 35.8 31.1 NA 0.03
Lineman or linebacker position
(vs. backfield position)
— no. of players (%)
27/53 (51) 5/5 (100) 22/48 (46) 10.6 (1.3–
∞
) 0.02
Surgery in past year
— no. of players (%)
16/53 (30) 3/5 (60) 13/48 (27) 3.5 (0.6–18.8) 0.13
Hospitalization in past year
— no. of players (%)
10/53 (19) 0/5 10/48 (21) 0.4 (0–3.6) 0.33
Use of antimicrobials in past year
— no. of players (%)
30/51 (59) 5/5 (100) 25/46 (54) 7.8 (0.5–
∞
) 0.06
Turf burns covered during games
— no. of players (%)
39/50 (78) 3/5 (60) 36/45 (80) 0.4 (0.1–2.2) 0.31
Shaved body other than face
— no. of players (%)
9/51 (18) 1/5 (20) 8/46 (17) 1.2 (0.2–9.2) 0.89
Gloves worn during games
— no. of players (%)
44/52 (85) 5/5 (100) 39/47 (83) 2.2 (0.2–
∞
) 0.41
Gloves used >3 times (vs. 1, 2,
or 3 times) before washing
— no. of players (%)
29/46 (63) 2/5 (40) 27/41 (66) 0.4 (0.1–2.1) 0.26
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3
,
2005
The
new england journal
of
medicine
474
colonization and transmission to other players before
our nasal-swab survey.
We found the highly conserved USA300-0114
MRSA clone was present in diverse regions of the
United States. This clone and other USA300 and
USA400 strains appear to have caused the majority
of community-associated MRSA cases characterized
to date in the United States.
17
The reasons for
the emergence of the clone are unclear; however,
antimicrobial use in the community may have helped
select bacteria that are resistant to standard empiric
therapy for skin and soft-tissue infections (i.e.,
a first-generation cephalosporin or a penicillinaseresistant
penicillin). The players in our investigation
were receiving 10 times the number of antimicrobial
prescriptions dispensed to the general
public. Increased use of antimicrobial agents, when
combined with other factors such as compromised
skin, close skin-to-skin contact, close person-toperson
proximity, a contaminated environment,
and suboptimal hand and personal hygiene may
provide the right conditions for efficient transmission
among the members of a cohort and thus lead
to clusters of skin infections.
On the basis of the findings of this and other investigations
of outbreaks among sports participants,
several recommendations can be made. First,
clinicians and other personnel involved in the care
of sports participants should be aware of the emergence
in the community of MRSA with distinct
microbiologic and epidemiologic characteristics.
Infections with these organisms predominantly
cause skin abscesses in otherwise healthy persons
who often have no health care exposures. Obtaining
cultures in suspected cases of infection and performing
antimicrobial-susceptibility testing will
facilitate early identification of cases and initiation
of targeted treatment. Clinicians should drain
abscesses and ensure that wounds are covered and
contained with clean, dry dressings. Infected persons
should receive guidance regarding enhanced
hand and personal hygiene to prevent transmission.
Frequently touched surfaces should be cleaned
in accordance with manufacturer-recommended
guidelines. Chlorhexidine-containing soap and nasal
decolonization with mupirocin have been recommended
to control outbreaks
19-23
; however, data
demonstrating the independent benefit of these
agents in controlling MRSA in community clusters
are lacking. Some studies have reported that anti-
Figure 3. Pulsed-Field Gel Electrophoresis Patterns with Percentage Similarities for MRSA Isolates from Competing Football Teams (St. Louis
Rams and Team A) and from Outbreaks in Various Community Settings and Geographic Locations.
The dendrogram presents only one representative MRSA isolate for each setting. All strains are from the staphylococcus PulseNet database
of the Centers for Disease Control and Prevention (CDC).
Similarity (%) Isolate Reference
St. Louis Rams professional football
team, Missouri
Professional football team A
College football team, California CDC19
CDC4
CDC4
CDC20
CDC21
CDC21
Martinez-Aguilar et al.22
CDC19
CDC19
College football team, Pennsylvania
Fencers, Colorado
Prison inmates, Mississippi
Prison inmates, Georgia
Prison inmates, Texas
Children, Tennessee
Children, California
Men who have sex with men, California
Hospital Strain USA100
Hospital Strain USA200
60 80 100
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.
n engl j med
352;5
www.nejm.org february
3, 2005
methicillin-resistant
s. aureus
among football players
475
bacterial soap with 1.5 percent triclocarban is effective
in preventing impetigo and atopic dermatitis.
24
Additional studies are needed to determine
whether the use of antibacterial soap should be routinely
recommended and whether decolonization
and the use of body antiseptics are also needed to
control transmission.
The CDC has initiated a collaboration with the
National Collegiate Athletic Association in developing
guidelines for the prevention and control of
community-associated MRSA among college football
players. The guidelines will include educational
materials targeted to athletic trainers and will
describe infection-control practices and measures
for responding to cases or clusters of infections.
To monitor the prevalence of community-associated
MRSA infections, the CDC has initiated active
population-based surveillance in eight geographic
locations in the United States.
25
These data will
help to characterize the emergence of MRSA in the
community and will guide public health interventions,
including strategies to prevent antimicrobial
resistance.
The use of trade names and commercial sources does not imply
endorsement by the U.S. Department of Health and Human Services
or the CDC.
We are indebted to Laura Rose, Terri Forster, Chesley Richards,
and Linda McCaig (CDC); to Dr. Bao-Ping Zhu (Missouri Department
of Health and Senior Services); and to Dr. William Baine
(Agency for Healthcare Research and Quality, U.S. Department of
Health and Human Services) for their support.
references
1.
Goodman RA, Thacker SB, Solomon
SL, Osterholm MT, Hughes JM. Infectious
diseases in competitive sports. JAMA 1994;
271:862-7.
2.
Dorman JM. Contagious diseases in
competitive sport: what are the risks? J Am
Coll Health 2000;49:105-9.
3.
Shopsin B, Mathema B, Martinez J, et al.
Prevalence of methicillin-resistant and methicillin-
susceptible
Staphylococcus aureus
in the
community. J Infect Dis 2000;182:359-62.
4.
Methicillin-resistant
Staphylococcus aureus
infections among competitive sports participants
— Colorado, Indiana, Pennsylvania,
and Los Angeles County, 2000–2003. MMWR
Morb Mortal Wkly Rep 2003;52:793-5.
5.
Naimi TS, LeDell KH, Como-Sabetti K,
et al. Comparison of community- and health
care-associated methicillin-resistant
Staphylococcus
aureus
infection. JAMA 2003;290:
2976-84.
6.
Chambers HF. The changing epidemiology
of
Staphylococcus aureus?
Emerg Infect
Dis 2001;7:178-82.
7.
2002 National Ambulatory Medical Care
Survey public use data file documentation.
Hyattsville, Md.: National Center for Health
Statistics, 2004.
8.
2002 National Hospital Ambulatory
Medical Care Survey public use data file documentation.
Hyattsville, Md.: National Center
for Health Statistics, 2004.
9.
Performance standards for antimicrobial
susceptibility testing. 14th informational
supplement M100-S14. Vol. 24. No. 1.
Wayne, Pa.: National Committee for Clinical
Laboratory Standards, January 2004.
10.
Jorgensen JH, Crawford SA, McElmeel
ML, Fiebelkorn KR. Detection of inducible
clindamycin resistance of staphylococci in
conjunction with performance of automated
broth susceptibility testing. J Clin Microbiol
2004;42:1800-2.
11.
Schreckenberger PC, Ilendo E, Ristow
KL. Incidence of constitutive and inducible
clindamycin resistance in
Staphylococcus aureus
and coagulase-negative staphylococci in
a community and a tertiary care hospital.
J Clin Microbiol 2004;42:2777-9.
12.
Katayama Y, Ito T, Hiramatsu K. A new
class of genetic element, staphylococcus cassette
chromosome mec, encodes methicillin
resistance in Staphylococcus aureus. Antimicrob
Agents Chemother 2000;44:1549-55.
13.
Lina G, Piemont Y, Godail-Gamot F, et
al. Involvement of Panton-Valentine leukocidin-
producing
Staphylococcus aureus
in primary
skin infections and pneumonia. Clin
Infect Dis 1999;29:1128-32.
14.
Letertre C, Perelle S, Dilasser F, Fach P.
A strategy based on 5' nuclease multiplex
PCR to detect enterotoxin genes sea to sej of
Staphylococcus aureus.
Mol Cell Probes 2003;
17:227-35.
15.
Maslow JN, Slutsky AM, Arbeit RD. Application
of pulsed-field gel electrophoresis
to molecular epidemiology. In: Persing DH,
Smith TF, Tenover FC, White TJ, eds. Diagnostic
molecular microbiology: principles
and applications. Washington, D.C.: American
Society for Microbiology, 1993:563-72.
16.
Tenover FC, Arbeit RD, Goering RV. How
to select and interpret molecular strain typing
methods for epidemiological studies of
bacterial infections: a review for healthcare
epidemiologists. Infect Control Hosp Epidemiol
1997;18:426-39.
17.
McDougal LK, Steward CD, Killgore
GE, Chaitram JM, McAllister SK, Tenover
FC. Pulsed-field gel electrophoresis typing
of oxacillin-resistant
Staphylococcus aureus
isolates from the United States: establishing
a national database. J Clin Microbiol 2003;
41:5113-20.
18.
Enright MC, Day NP, Davies CE, Peacock
SJ, Spratt BG. Multilocus sequence typing
for characterization of methicillin-resistant
and methicillin-susceptible clones of
Staphylococcus aureus.
J Clin Microbiol 2000;
38:1008-15.
19.
Outbreaks of community-associated
methicillin-resistant
Staphylococcus aureus
skin
infections — Los Angeles County, California,
2002–2003. MMWR Morb Mortal Wkly
Rep 2003;52:88.
20.
Methicillin-resistant
Staphylococcus aureus
skin or soft tissue infections in a state prison
— Mississippi, 2000. MMWR Morb Mortal
Wkly Rep 2001;50:919-22.
21.
Methicillin-resistant
Staphylococcus aureus
infections in correctional facilities —
Georgia, California, and Texas, 2001–2003.
MMWR Morb Mortal Wkly Rep 2003;52:
992-6.
22.
Martinez-Aguilar G, Avalos-Mishaan A,
Hulten K, Hammerman W, Mason EO Jr,
Kaplan SL. Community-acquired, methicillin-
resistant and methicillin-susceptible
Staphylococcus aureus musculoskeletal infections
in children. Pediatr Infect Dis J 2004;
23:701-6.
23.
Baggett HC, Hennessy TW, Rudolph K,
et al. Community-onset methicillin-resistant
Staphylococcus aureus associated with
antibiotic use and the cytotoxin Panton-Valentine
leukocidin during a furunculosis outbreak
in rural Alaska. J Infect Dis 2004;189:
1565-73.
24. Luby S, Agboatwalla M, Schnell BM,
Hoekstra RM, Rahbar MH, Keswick BH. The
effect of antibacterial soap on impetigo incidence,
Karachi, Pakistan. Am J Trop Med Hyg
2002;67:430-5.
25. Hageman J, Morrison M, Como-Sabetti
K, et al. Multiple state-based surveillance
for community-associated methicillin-resistant
Staphylococcus aureus (CAMRSA) disease:
Maryland, Minnesota, and Georgia 2001-
2002. Presented at the International Conference
on Emerging Infectious Diseases, Atlanta,
February 29–March 3, 2004. abstract.
Copyright © 2005 Massachusetts Medical Society.
Downloaded from www.nejm.org on October 22, 2006 . Copyright © 2005 Massachusetts Medical Society. All rights reserved.

Health and safety implications of injury in
professional rugby league football
C. Gissane1, J.White2, K. Kerr3, S. Jennings4 and D. Jennings5
Background Professional sport is characterized by high injury rates but is also covered by health
and safety legislation.
Aim To examine the incidence of injury in professional rugby league as defined by the
Reporting of Diseases and Dangerous Occurrences Regulations 1995 (RIDDOR
95).
Methods All injuries received during playing and training to both first-team and ‘academy’
(<19 years old) players during two playing seasons at one professional club were
recorded. The length of time a player was unable to take part in full training and
playing was used as a measure of severity. Injuries were classified into minor injuries
(0–3 days), over 3 day injuries or major injuries, in which the final two categories
corresponded with RIDDOR 95.
Results Thirty-two per cent (95% confidence interval = 26–39%) of all injuries received
satisfied the RIDDOR 95 criteria. The overall injury rate was 8.5 per 1000 h
(7.2–9.9) for the first team and 4.1 per 1000 h (3.2–5.4) for the academy team.
During match play the first-team injury rate was 157.7 per 1000 h (133.5–185.1)
and 67.7 (51–81.1) for the academy team.Training injury rates were lower, at 0.5 per
1000 h (0.2–1.0) and 0.3 per 1000 h (0.1–0.8), respectively.
Conclusion The injury rate for professional rugby league is much higher than reported in other
high-risk occupations such as mining and quarrying. The large differences in injury
rates between first and academy teams have implications for young players likely to
progress to first-team status.
Key words Health and safety; high injury rates; high-risk occupations; rugby league.
Received 6 August 2002
Revised 19 March 2003
Accepted 20 June 2003
Introduction
Professional athletes are subject to health and safety law
in the same way as other employed persons at work [1].
This places a legal duty on sports employers to assess the
risks associated with the sporting activity and reduce
these risks to an acceptable level [1,2]. As part of this
legal obligation, it is necessary to carry out health
surveillance when adverse health conditions are related to
the working practices, so that surveillance may serve to
facilitate the protection of the health of employees [3].
The Reporting of Diseases and Dangerous Occurrences
Regulations 1995 (RIDDOR 95) require that certain
injuries are reported to a designated authority, such as
the Health & Safety Executive (HSE) or local authority,
and that records are kept of all injuries at the place of
employment [4].
Professional rugby league is a contact sport that
involves physical collisions between players. A rugby
Occupational Medicine, Vol. 53 No. 8
© Society of Occupational Medicine 2003; all rights reserved 512
1Department of Health and Social Care, Brunel University, Osterley Campus,
Isleworth, Middlesex TW7 5DU, UK.
2Centre for Sports Medicine, University of Nottingham, Queen’s Medical
Centre, Nottingham NG7 2UH, UK.
3Division of Physiotherapy Education, City Hospital, Hucknall Road,
Nottingham NG5 1PG, UK.
4Shoulder Unit, Royal National Orthopaedic Hospital, Stanmore, Middlesex
HA7 4LP, UK.
5Wall House Surgery, Yorke Road, Reigate, Surrey, RH2 9HG, UK.
Correspondence to: C. Gissane, Department of Health and Social Care,
Brunel University, Osterley Campus, Isleworth, Middlesex TW7 5DU, UK.
e-mail: conor.gissane@brunel.ac.uk
Occupational Medicine 2003;53:512–517
DOI: 10.1093/occmed/kqg103
league team consists of 13 players (six forwards and seven
backs); each team has sets of six tackles to advance the
ball down field. The ball must be passed backwards, but
can be carried or kicked down field. Unlike some sports,
such as American football, there are no special teams or
sub-units within a team (other than forwards and backs),
so each player has a role to play in both attack and
defence. Each team has four replacements, and a team is
allowed to make a maximum of six substitutions during a
game, so a player may leave the field and return at a later
point in the match.
Previous research into the incidence of injury in rugby
league has shown that the injury rates appear to be higher
than in many other sports [5–14]. It has also been
demonstrated that forwards are at a greater risk of injury
than backs [6,8,10,14], and the tackled player carrying
the ball is more likely to be injured in a collision than a
player who is a tackler [6,10,14]. There is also evidence
to suggest that retired players often suffer long-term
consequences of accumulated injuries received during
their playing careers, such as chronic back pain, restricted
joint mobility and inability to work in their chosen
occupations [15].
However, comparisons among injury studies are
sometimes difficult because of the various definitions
used of what constitutes a sports injury. For example,
some studies have used the definition of unavailable for
selection for the next competitive game [7,9,10], while
others have included all injuries that require attention
by the medical team [6,8,12,14]. The stated purpose
of RIDDOR 95 is to provide data that will indicate how
and where risks arise with a view to helping prevent
injuries, ill-health and loss of time away from work due
to accidents. The purpose of the present study was to
determine the incidence of injury in professional rugby
league as defined by the RIDDOR 95 regulations, in
terms of major injuries and over 3 day (o3d) injuries [4].
Methods
The subjects in the study were the players at one
professional rugby league club who were monitored over
a period of two seasons, and included both the first and
academy (under 19 years) teams. The club in question
was one of 14 in the European Super League. Each injury
received was diagnosed by either the club doctor or
the club physiotherapist. The data recorded about each
specific injury were the date that the injury occurred, a
description of the injury and the activity at the time of the
injury. For the purposes of the present study, an injury
was defined as any injury that was received during either
playing or training (time at work) and that required
attention by one of the medical team. The length of time
that a player was unable to take part in full training and
playing was used as a measure of severity. Injuries were
classified into minor (0–3 days), o3d or major injuries,
the last two categories corresponding to reportable
injuries under RIDDOR 95 [4]. Major injuries were
defined as: fractures other than to fingers, toes and
thumbs; dislocations of the shoulder, hip, knee or spine;
any other injury leading to heat-induced illness or
unconsciousness, or requiring resuscitation; or any injury
requiring admittance to hospital for >24 h [4]. o3d
injuries are those injuries that prevent a player from
carrying out his normal duties for a period >3 working
days [4].
During the first season there were 36 first-team players
and 27 academy players, while during the second season
of observation there were 34 first-team players and
28 academy team players available for the study. The
time exposed to risk during competitive matches was
calculated as the number of players multiplied by 1.33 h
(the duration of the game) multiplied by the number of
games played, which resulted in 1764 playing hours. The
time exposed to risk during training was calculated as the
number of players multiplied by the average number of
hours of training per week, which totalled 31 248 h. The
number of exposure (playing and training) hours totalled
33 011.
Statistical analysis
Injury frequency rates (IFR) per 1000 h and injury
incidence rates (IIR) per 1000 players were calculated
using the methods described by the HSE [16].
To decide if differences between IFRs were significant,
the method described by Clarke [17] to compare person
time rates was used.
Results
During the observation period of this study, there were a
total of 218 injuries recorded. Thirty-two per cent of all
injuries reported satisfied the criteria to be reportable
under RIDDOR 95 [95% confidence interval (CI) =
26–39%].
Injury frequency rates
The numbers and IFRs of first and academy team players
during overall, playing and training exposure are
displayed in Table 1. The overall, playing and training
IFRs were approximately twice as high among the first
team when compared with the academy team. There
IFR =
No. of injuries during the period
Total hour s worked during the period
F
H GI K J
×100
IIR =
No. of injuries during the period
Mean no. of employees during the period
F
H G
I
K J
×100
C. GISSANE ET AL.: HEALTH AND SAFETY IMPLICATIONS OF RUGBY LEAGUE INJURIES 513
were significant differences between the first and
academy teams for overall injuries (difference = 4.31 per
1000 h, z = 5.01, P < 0.001) and playing injuries
(difference = 90.1 per 1000 h, z = 5.7, P < 0.001), but
not for training injuries (difference = 0.13 per 1000 h,
z = 0.56, P = 0.58).
The first team demonstrated an overall higher number
of injuries in each of the categories reported. During
playing, there was a similar situation, with first-team
players recording more injuries in each category than
academy players, while during training there were few
reportable injuries (Table 2). There were significant
differences between first and academy teams in the
respective IFRs for o3d injuries (difference = 1.52 per
1000 h, z = 3.61, P < 0.001), reportable injuries
(difference = 1.78 per 1000 h, z = 3.57, P < 0.001),
o3d injuries received while playing (difference = 29.58,
z = 3.76, P < 0.001) and reportable injuries received
while playing (difference = 33.76 per 1000 h, z = 3.7,
P < 0.001).
Injury incidence rates
The numbers and IIRs among the three categories of
injury for the first and academy teams are displayed
in Table 2. The overall, playing and training IIRs were
approximately twice as high among the first team when
compared with the academy team. There were significant
differences between the first and academy teams for o3d
injuries (difference = 412 per 1000 players, z = 4.52,
P < 0.001) and reportable injuries (difference = 462 per
1000 players, z = 5.05, P < 0.001).
The injury types and severity categories among
first-team and academy players are presented in Table 3.
Overall in the minor injury category, four types of injury
accounted for the high totals of all injuries received by
first-team (79/105) and academy players (34/42). However,
there were differences in the profile of the injury type
among first-team players, with haematoma the most
common, followed by joint sprain, muscle strain and
laceration, compared with academy players, for whom
joint sprain was the most common, followed by con-
Table 1. Injury frequency rates (per 1000 h) for first and academy teams
First team Academy team
n IFR 95% CI n IFR 95% CI
Overall 159 8.5 7.2–9.9 59 4.1 3.2–5.4
Minor 105 5.6 4.6–6.8 42 3.0 2.1–4.0
o3d 43 2.3* 1.7–3.1 11 0.8 0.4–1.4
Major 11 0.6 0.3–1.0 6 0.4 0.2–0.9
No. reportable 54 2.9* 2.2–3.7 17 1.2 0.7–1.9
Playing 150 157.7 133.5–185.1 55 67.7 51.0–88.1
Minor 98 103.1 83.7–125.6 38 46.8 33.1–64.2
o3d 41 43.1* 30.9–58.5 11 13.5 6.8–24.2
Major 11 11.6 5.8–20.7 6 7.4 2.7–16.1
No. reportable 52 54.7* 40.8–71.7 17 20.9 12.2–33.5
Training 9 0.5 0.2–1.0 4 0.3 0.1–0.8
Minor 7 0.4 0.2–0.8 4 0.3 0.1–0.8
o3d 2 0.1 0.0–0.4 0 0.0
Major 0 0.0 0 0.0
No. reportable 2 0.1 0.0–0.4 0 0.0
*Significantly different between first and academy teams.
Table 2. The number of injuries and injury incidence rates for first and academy teams
First team Academy team
n IIR per 1000
players
95% CI n IIR per 1000
players
95% CI
Minor 105 1479 1209–1790 42 737 531–995
o3d* 43 606 438–815 11 193 96–345
Major 11 155 77–277 6 105 38–229
No. reportable* 54 761 571–992 17 298 173–477
*Significantly different between first and academy teams.
514 OCCUPATIONAL MEDICINE
tusions, haematomas and lacerations. Unfortunately, the
numbers of injuries were not large enough to permit
statistical analysis.
In the o3d injury category, joint sprain, muscle strains
and haematomas accounted for the majority of injuries in
first-team players (36/43) and all injuries in academy
players. Furthermore, the profile of injury type was
similar for both player groups with joint sprain the most
common, followed by muscle strain and haematoma.
In the major injury category, fractures and dislocations
accounted for most of the injuries received by first-team
and academy players, with concussion the other category
of injury received by both player groups.
The median time lost among all players as a result of
being injured was none for minor injuries, 11 (interquartile
range = 4–46) days for o3d injuries and 24
(0–47) days for major injuries. Furthermore, there were
no significant differences between the first and academy
team players in median days off for o3d (12.5 versus
9.5 days, z = 1.44, P = 0.15), nor major injuries (25
versus 11 days, z = 0.68, P = 0.51).
Mechanism of injury
Being tackled was the activity resulting in 44% of all
injuries incurred, and the figures for minor, o3d and
major were 38, 46 and 47%, respectively. Tackling a ball
carrier was associated with 30% of all injuries, and the
figures for minor, o3d and major injuries were 31, 17
and 41%, respectively. Injuries that occurred away from
the tackle accounted for 26% of the total injuries, and
included such incidences as running, changing direction
and turning, as well as a very small number of scrum
injuries.
Discussion
The major findings of this study were the very high injury
rates that were observed with an IIR for reportable
injuries of 563 per 1000 players, which compares with a
relatively low rate of 32 per 1000 employees for workers
in dangerous occupations such as mining and quarrying
of energy producing materials [18]. There could be
several reasons for this difference. First, in the present
study, the authors attempted to record all injuries that
took place, which could result in the authors paying
special attention to the details relating to injury occurrence
in a group of workers who have previously not been
the subject of such investigation.
Secondly, since the HSE is only able to report on
injuries that are reported to the relevant authorities,
this leaves the possibility that some injuries may go
unreported. In support of this conjecture, it has been
suggested that injury rates from the Labour Force Survey
[19] may be higher than those reported to the HSE [20].
Labour Force Survey figures are based on a combination
of employer reports, social security and insurance claims,
of which the latter have a relatively low level of underreporting
[20].
However, high injury rates in sport are not uncommon.
The IFR for professional soccer was reported to
be around 1000 times higher than for other industrial
occupations (e.g. construction and mining) that are
considered to be high risk. However, in that study [2], the
IFR for match play was considerably lower than that
found in the present study (27.7 versus 115 per 1000 h).
The overall IFR reported for professional soccer was
slightly higher than in the present study (8.5 versus 6.6
per 1000 h), and the figure for training was eight times
higher (3.5 versus 0.42 per 1000 h). In the present study,
only time training in running activities was included, with
other training methods, such as swimming and weight
training, being excluded. Additionally, coaches in rugby
league might be extremely reluctant to replicate match
conditions exactly in training because of the high injury
risks associated with collisions in the tackle. If the time
practising these techniques does not occupy a large
proportion of training time, this could account for the
differences in IFRs between playing and training. Rugby
league coaches are also required to examine the coaching
system for possible dangers and monitor the players for
safety in their practices, and while this is aimed primarily
Table 3. Injuries by type and severity category for the first team
and the academy team
First Academy
n % n %
Minor
Muscle strain 18 17 3 7
Haematoma 23 22 5 12
Contusion 10 10 6 14
Joint sprain 21 20 18 43
Concussion 7 7 3 7
Laceration 17 16 5 12
Fracture and dislocation 6 6 1 2
Abrasion 1 1 1 2
Other 2 2 0 0
Total 105 42
o3d
Muscle strain 14 33 3 27
Haematoma 5 12 2 18
Contusion 1 2 0 0
Joint sprain 17 40 6 55
Concussion 2 5 0 0
Fracture and dislocation 3 7 0 0
Other 1 2 0 0
Total 43 11
Major
Fracture and dislocation 10 91 5 83
Concussion 1 9 1 17
Total 11 6
C. GISSANE ET AL.: HEALTH AND SAFETY IMPLICATIONS OF RUGBY LEAGUE INJURIES 515
at child participation, it is a theme permeating all age
groups and levels.
It was interesting to note the difference in injury rates
between the first and academy teams. The IFRs for both
training and playing were lower for the academy team in
each category of injury, which is in agreement with
previous studies [21]. This may have considerable
implications for academy players who, while still young
enough to play for the academy team, are considered
to be of first-team standard. It would be interesting
to compare the injury rates experienced by 18- and
19-year-old players playing first-team rugby league with
those of their older counterparts.
In the present study, 32% of injuries were classified as
reportable injuries, which is a lower proportion than
reported for professional soccer (84%) [2]. In addition,
the median number of days lost due to reportable injury
in the present study (8 days) was lower than the number
of days lost in soccer (14.6 days). However, when minor
injuries in rugby league are excluded, the median figure is
much closer at 12 days.
The finding that fractures represent the greatest
proportion of non-fatal major injuries is in line with the
figures of the HSE [18]. However, the figure in the
current study is slightly higher than the HSE estimate
[88 (95% CI 65–96) versus 76%]. Also in common with
the HSE, strains and sprains constituted the most
common specific causes of o3d injuries [18], but the
figures in the present study were considerably higher than
those reported by the HSE (87 versus 42%). A possible
reason for these differences might be the very peculiar
nature of sports injuries. It has been suggested that if a
person is sick or injured, they may want to take time away
from work, but if a person is injured playing sport, they
often want to continue playing, or return to play before
they have fully recovered from the injury [22]. This may
be true whether or not the player is a professional sportsperson,
which may well have the effect of reclassifying
what might on first inspection be o3d injuries, into the
minor injuries categories and distorting the findings. For
example, in the present study, there were three instances
of nasal fractures for which the player did not stop playing
or training. RIDDOR 95 does not exclude them but they
resulted in no time away from work (playing rugby).
This explains the anomaly of the inter-quartile range of
0–47 days in this category.
The finding that the tackle is the activity that is
associated with most injury has been reported previously
in rugby league [6,8–10] and also in rugby union [23]. In
recent years, attempts have been made with rule changes,
in order tomake certain tackles less dangerous [14]. It has
further been suggested that, due to the nature of the
tackle, it might be time to review the procedures and
practices in both codes of the game [24].
Professional sport is covered by health and safety
regulations, and rugby has been described as a serious
source ofmorbidity in young men [23].The present study
has demonstrated that for professional rugby league
injury rates are much higher than for some high-risk
occupations, and higher during match play compared
with training. If risk assessments are to be carried out in
an effort to reduce the risk of injury, any health
surveillance needs to be able to examine the activities
during playing time, as the risk of injury in this phase of
players’ work is much greater than during training.
Successful health and safetymanagement should focus on
an organization’s most valuable asset: its workers [16]. In
professional sport, the workforce is highly skilled and
relatively expensive to employ; therefore, as in any other
profession, it is important to assess the risk to injury so
that it can possibly be reduced.
References
1. Fuller C. Implications of health and safety legislation
for the professional sports person. Br J Sports Med
1995;29:5–9.
2. Hawkins R, Fuller C. A prospective epidemiological study
of injuries in four English professional football clubs. Br J
Sports Med 1999;33:196–203.
3. Health & Safety Council (HSC). Management of Health and
Safety at Work: Approved Code of Practice. London: HMSO,
1992.
4. HSE. A Guide to the Reporting of Injuries, Diseases and
Dangerous Occurrences Regulations 1995. London: HMSO,
1996.
5. Gabbett T. Incidence of injury in semi-professional rugby
league players. Br J Sports Med 2003;37:36–44.
6. Gissane C, Jennings D, Standing P. Incidence of injury in
rugby league football. Physiotherapy 1993;79:305–310.
7. Gibbs N. Injuries in professional rugby league: a three-year
prospective study of the South Sydney professional rugby
league football club. Am J Sports Med 1993;21:696–700.
8. Gissane C, Jennings D, Cumine A, Stephenson S, White J.
Differences in the incidence of injury between rugby league
forwards and backs. Aust J Sci Med Sport 1997;29:91–94.
9. Gissane C, Phillips LH, Jennings D, White JA, Cumine A.
Injury in rugby league: the new super league. Br J Sports
Med 1997;31:85.
10. Gissane C, Jennings D, White J, Cumine A. Injury in
summer rugby league football: the experiences of one club.
Br J Sports Med 1998;32:149–152.
11. Gissane C, Jennings D, Kerr K, White J. A pooled data
analysis of injury incidence in rugby league football. Sports
Med 2002;32:211–216.
12. Hodgson-Phillips L, Standen P, Batt M. Effects of seasonal
change in rugby league on the incidence of injury. Br J
Sports Med 1998;32:144–148.
13. Seward H, Orchard J, Hazard H, Collinson D. Football
injuries in Australia at the elite level. Med J Aust
1993;159:298–301.
14. Stephenson S, Gissane C, Jennings D. Injury in rugby
516 OCCUPATIONAL MEDICINE
league: a four year prospective survey. Br J Sports Med
1996;30:341–345.
15. Meir R, McDonald K, Russell R. Injury consequences
from participation in professional rugby league: a preliminary
report. Br J Sports Med 1997;33:132–134.
16. HSE. Successful Health and Safety Management. London:
HMSO, 1992.
17. Clarke G. Statistics and Experimental Design: an Introduction
for Biologists and Biochemists. London: Edward Arnold,
1994.
18. HSC. Health and Safety Statistics: Statistical Supplement to
the 1996/97 Annual Report. London: HMSO, 1997.
19. International Labour Organization (ILO). Yearbook of
Labour Statistics 1999. Geneva: ILO, 1999.
20. HSE. Workplace Injury: Comparison of Great Britain and
Europe and the USA. London: HMSO, 1997.
21. Estell J, Shenstone B, Barnsley L. Frequency of injuries in
different age groups in an elite rugby league club. Aust J Sci
Med Sport 1995;27:95–97.
22. Schootman M, Powell JW, Albright JP. Statistics in sports
injury research. In: Delee JC, Drez D, eds. Orthopaedic
Sports Medicine. Philadelphia, PA: WB Saunders, 1994;
160–183.
23. Garraway M, MacLeod D. Epidemiology of rugby football
injuries. Lancet 1995;345:1485–1487.
24. Milburn P. The rugby tackle—a time for review. J Physic
Educ N Z 1995;28:9–15.
C. GISSANE ET AL.: HEALTH AND SAFETY IMPLICATIONS OF RUGBY LEAGUE INJURIES 517

doi:10.1136/bjsm.2004.014571
Br. J. Sports Med. 2005;39;542-546
M Waldén, M Hägglund and J Ekstrand
2001–2002 season
study of injuries in professional football during the
UEFA Champions League study: a prospective
http://bjsm.bmjjournals.com/cgi/content/full/39/8/542
Updated information and services can be found at:
These include:
References
http://bjsm.bmjjournals.com/cgi/ ... 8/542#otherarticles
1 online articles that cite this article can be accessed at:
http://bjsm.bmjjournals.com/cgi/content/full/39/8/542#BIBL
This article cites 23 articles, 10 of which can be accessed free at:
Rapid responses
http://bjsm.bmjjournals.com/cgi/eletter-submit/39/8/542
You can respond to this article at:
service
Email alerting
top right corner of the article
Receive free email alerts when new articles cite this article - sign up in the box at the
Topic collections
• Sports Medicine (1329 articles)
• Injury (833 articles)
• Other Epidemiology (1667 articles)
Articles on similar topics can be found in the following collections
Notes
http://www.bmjjournals.com/cgi/reprintform
To order reprints of this article go to:
http://www.bmjjournals.com/subscriptions/
To subscribe to British Journal of Sports Medicine go to:
Downloaded from bjsm.bmjjournals.com on 22 October 2006
ORIGINAL ARTICLE
UEFA Champions League study: a prospective study of
injuries in professional football during the 2001–2002
season
M Walde´n, M Ha¨gglund, J Ekstrand
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
See end of article for
authors’ affiliations
. . . . . . . . . . . . . . . . . . . . . . .
Correspondence to:
Dr Walde´n, Department of
Health and Society,
Linko¨ping University,
S-581 83 Linko¨ping,
Sweden; markus.walden@
telia.com
Accepted
23 November 2004
. . . . . . . . . . . . . . . . . . . . . . .
Br J Sports Med 2005;39:542–546. doi: 10.1136/bjsm.2004.014571
Background: No previous study on adult football involving several different countries has investigated the
incidence and pattern of injuries at the highest club competitive level.
Objective: To investigate the risk exposure, risk of injury, and injury pattern of footballers involved in UEFA
Champions League and international matches during a full football season.
Method: Eleven top clubs (266 players) in five European countries were followed prospectively throughout
the season of 2001–2002. Time-lost injuries and individual exposure times were recorded during all club
and national team training sessions and matches.
Results: A total of 658 injuries were recorded. The mean (SD) injury incidence was 9.4 (3.2) injuries per
1000 hours (30.5 (11.0) injuries per 1000 match hours and 5.8 (2.1) injuries per 1000 training hours).
The risk of match injury was significantly higher in the English and Dutch teams than in the teams from
France, Italy, and Spain (41.8 (3.3) v 24.0 (7.9) injuries per 1000 hours; p = 0.008). Major injuries
(absence .4 weeks) constituted 15% of all injuries, and the risk of major injury was also significantly
higher among the English and Dutch teams (p = 0.04). National team players had a higher match
exposure, with a tendency towards a lower training injury incidence than the rest of the players (p =
0.051). Thigh strain was the most common injury (16%), with posterior strains being significantly more
common than anterior ones (67 v 36; p,0.0001).
Conclusions: The risk of injury in European professional football is high. The most common injury is the
thigh strain typically involving the hamstrings. The results suggest that regional differences may influence
injury epidemiology and traumatology, but the factors involved are unclear. National team players have a
higher match exposure, but no higher risk of injury than other top level players.
Accor