Scand J Med Sci Sports 2010: 20: 588–592
doi: 10.1111/j.1600-0838.2009.00995.x
& 2009 John Wiley & Sons A/S
Acute hamstring injuries in Danish elite football: A 12-month
prospective registration study among 374 players
J. Petersen1, K. Thorborg1, M. B. Nielsen2, P. Hölmich1
1
Department of Orthopaedic Surgery, Amager Hospital, Faculty of Health Sciences, University of Copenhagen, Copenhagen,
Denmark, 2Department of Radiology, Section of Ultrasound, Diagnostic Centre, Rigshospitalet, Faculty of Health Sciences,
University of Copenhagen, Copenhagen, Denmark
Corresponding author: Jesper Petersen, Department of Orthopaedic Surgery, Amager Hospital, Italiensvej 1, Copenhagen, DK
2300, Denmark. Tel: 145 3234 3267, Fax: 145 3234 3995, E-mail: [email protected]
Accepted for publication 8 Jun 2009
The purpose of the present study was to examine the
incidence rates of acute hamstring injuries in Danish elite
football sustained during training or match play. Furthermore, it was our intention to document details about the
recurrence, severity and the injury seasonal distribution.
Hamstring injuries among 374 elite football players were
registered prospectively during a 12-month period. A total
of 46 first-time and eight recurrent hamstring injuries were
registered. The incidence rates for incurring a first-time
hamstring injury showed a significantly (Po0.01) greater
incidence rate per 1000 h during match play compared with
training. Of 32 players who reported a hamstring injury in
the 12-month period before the study, eight (25%) incurred
an injury that fulfilled the criteria for a recurrent injury. In
69% of the injuries, the severity of injury was categorized as
moderate (8–28 days from injury to injury free) and 18% as
severe (428 days from injury to injury free). Each team
sustained a mean of 3.4 hamstring injuries per season, with a
mean of 21.5 days missed per injury (range 3–136; median
16 days per injury). The seasonal distribution showed an
accumulation of injuries in the first 2 months after a 3.5month mid-season winter break.
Injuries to the hamstring muscles are common within
several sports including the different types of football. The majority of hamstring injuries occur while
the athletes are running or sprinting (Arnason et al.,
1996; Woods et al., 2004). In elite football (soccer),
12–16% of all injuries are hamstring strains, which
make this injury the most prevalent injury type
(Arnason et al., 1996; Hawkins et al., 2001). The
incidence in football has been reported to be as high
as 4.1 injuries per 1000 match hours (Hägglund et al.,
2006). In addition to the high incidence, a common
problem concerning this injury is the high risk of
recurrence (Arnason et al., 1996; Woods et al., 2004).
The main purpose of the present study was therefore to examine the incidence rate of first-time hamstring injuries in Danish elite football sustained
during training or match play in a 12-month period.
Furthermore, it was our intention to document details about the recurrence, severity, characteristics of
injured players and the injury seasonal distribution.
football league, with the other five teams playing in the second
division competition.
The Danish ethics committee of the Capital Region and the
Danish Data Protection Agency approved the study.
Materials and methods
Subjects
A total of 374 Danish elite football players from 16 teams
committed themselves to participate in this study. At the
beginning of the study 11 teams played in the premier Danish
588
Inclusion
Only players in the first team squad were included in the study.
Data for players who played within one of the 16 teams at the
start of the study period of 12 months were included in the
prospective data analysis.
Exclusion
New players who joined a team in the study period were not
included in the study. Players who left a team before the end of
the study period were excluded if they did not remain within
one of the other participating teams. Players who sustained
hamstring injuries were excluded from further participation in
the study when they had returned to full participation in
training sessions and were available for match selection. Data
from the start of the study until the time for exclusion were
used in the incidence analysis.
Study period
The playing season in Denmark lasts for 10–11 months. The
prospective injury registration covered a full 12-month period
i.e. the second phase of the playing season (March–May/
June), preseason (June–July), the first phase of the playing
season (July–December) and the mid-season winter break
Acute hamstring injuries in Danish elite football
(December–March). The registration began with the start of
the second phase of the playing season (March 2007) and
ended 12 months later (March 2008).
Time of team-based and individual physical activities that
aimed at maintaining or improving players’ football skills or
physical condition and was under the control or the guidance
of the team’s coaching or fitness staff was used to calculate
training exposure.
Injury and baseline registration
At the beginning of the study, all players (n 5 374) reported
baseline information using an information form designed for
this study, and in accordance with the recommendation from
the Injury Consensus Group under the auspices of the FIFA
Medical Assessment and Research Centre (Fuller et al., 2006).
The data included in the analysis were the playing position,
age and details of hamstring injuries during the last 12 months.
The team physiotherapist and/or physician recorded all
hamstring injuries during the registration period using a
specific injury registration form designed for this study.
Contusion injuries to the posterior thigh were excluded.
Furthermore, the time until the injured player could return
to full participation was registered.
Exposure
The club physiotherapist in the participating teams made a
specification of the training exposure for the entire team and
the average number of players who participated. From these
data the individual training exposure was calculated. The first
author registered the total match exposure regarding all
league, cup and international matches for the participating
teams. From these data the match exposure was calculated for
the individual player.
Match and training exposure were calculated only for
players who did not report any hamstring injuries 12 months
before this study, in order to report the incidence rate of new
hamstring injuries only.
Definitions
Apart from the recurrence definition, all definitions in this
study were in accordance with the recommendation from the
FIFA Injury Consensus Group (Fuller et al., 2006).
An acute hamstring injury was defined as ‘‘any suddenly
occurred physical complaint in the region of the posterior
thigh sustained during a football match or training, irrespective of the need for medical attention or time-loss from
football activities.’’
A player was defined as injured until he returned to full
participation in team training and was available for match
selection. The medical staff, in consultation with the player,
assessed when an injured player could return to full participation.
For players who had a recalled hamstring injury in the last
year before this study, the definition for a recurrent injury was
‘‘an posterior thigh injury of the same type and at the same site
as the index injury and which occurred after the player had
returned to full participation from the index injury.’’
Data of previous hamstring injuries were used to assess
whether an injury sustained in the prospective study should be
categorized as a new or a recurrent injury. This made it
possible to calculate a recurrence rate for the subsequent
playing season.
The number of days that elapsed from the date of injury
(day zero) to the date of the player’s return to full participation in team training and availability for match selection was
used to calculate the injury severity. Injury severity was
divided into slight (0 day), minimal (1–3 days), mild (3–7
days), moderate (8–28) and severe (428 days).
The time of playing between teams from different clubs was
used to calculate match exposure.
Analysis
Injury incidences were calculated separately for match and
training exposure. The incidences are reported per 1000
player-hours and calculated using the formula: number of
injuries 1000/exposure hours. 95% confidence intervals (CI)
for match and training incidence rates were calculated.
SAS computer software was used for the analyses. Significance was set at Po0.05.
Results
Of the 374 players, 242 were followed for the entire
period of 12 months. Eighty-four players were followed for 4 months (second phase of the playing
season) and 48 players were followed for 10 months
(preseason, first and second phases of the playing
season). The reasons for dropout during the study
were transfer or stop of active career in the pre or the
mid-seasonal break.
Division, playing position and age distribution of
the cohort at the beginning of the study are presented
in Table 1.
Of the 374 players in the study 45 (12%) reported
50 hamstring injuries within 12 months before the
prospective study.
In the prospective study, a total of 46 first-time
hamstring injuries were recorded during the observation period of 12 months. The numbers of new
injuries in individual teams were 1–5, mean 2.9,
median 3.
Of all new injuries, 28 occurred during match play
and 14 during training. The circumstances for four
injuries were unknown due to recording error. A
total of 15 367 match hours and 118 644 training
hours were recorded. Incurring a first-time hamstring
injury showed a significantly greater incidence rate
Table 1. Division, playing position and age distribution of the cohort at
the beginning of the study
Division No. %
Premier 270
league
First
104
division
Playing
position
72 Goalkeeper
28 Defender
Midfielder
Forward
Alternating
Total
No. %
Age
No. %
distribution
40
10 17–22
163
44
104
28 23–28
137
36
63
6
5
17
2
1
130
68
29
35 29–34
18 351
8 Not
specified
Not specified
3
1
374 100 Total
374 100 Total
374 100
589
Petersen et al.
per 1000 h during match play compared with training
1.82 (95% CI 1.26–2.64) vs 0.12 (95% CI 0.07–0.20).
Of the 45 players reporting minimum one hamstring injury in the 12-month period before the
prospective study, 32 players were followed during
the entire 12-month study period. These 32 players
sustained a total of 8 injuries (25%) that fulfilled the
criteria for a recurrent injury.
The 46 new and eight recurrent injuries resulted in
a total of 1163 days of absence from football (range
3–136; mean 21.5; median 16 days per injury). The
injury severity is presented in Table 2. Each team
sustained a mean of 3.4 hamstring injuries per season
(range 1–5; median 3), resulting in an average of 73
days missed due to hamstring injury per team.
Table 3 shows the distribution of hamstring injuries by playing position and age.
Figure 1 shows the seasonal distribution of the 54
injuries during the 12-month registration period.
Discussion
This study shows that first-time and recurrent hamstring injuries constitute a significant problem in
Danish elite football. Furthermore, this study highlights match play as a high risk activity compared
with training, which has been shown in previous
football studies as well. Overall, other studies have
found injury incidences from 2.03 to 4.1 injuries per
1000 match hours and 0.25 to 0.7 injuries per 1000
training hours (Ekstrand & Gillquist, 1983; Arnason
Table 2. Injury severity among 374 Danish elite football players. Classification according to the number of days from injury to return to full
participation
Injury severity
No.
%
Slight (0 day)
Minimal (1–3 days)
Mild (4–7 days)
Moderate (8–28 days)
Severe (428 days)
Total
0
1
6
37
10
54
0
2
11
69
18
100
et al., 1996, 2004; Ekstrand et al., 2006; Hägglund
et al., 2006). However, it is important to underline
that the incidence rate in the present study is for new
injuries only. In contrast to studies concerning Australian Rules football (Orchard & Seward 2002;
Orchard et al., 2005), no football studies have used
a distinction between new and recurrent hamstring
injuries when reporting incidence rates. Thus, to our
knowledge, this is the first study that examines the
incidence of first-time hamstring injuries in elite male
football with a data registration period of 12 months,
with most other studies registering new and recurrent
injuries for the competitive season only (Arnason
et al., 1996, 2004; Hägglund et al., 2006). It is difficult
to perform inter-study comparisons of results if
different study designs, definitions and registration
periods are used. We have therefore chosen to follow
the recommendations by the Injury Consensus
Group (Fuller et al., 2006), and have registered
injuries both for the competitive and the preseason
in order to cover the injury incidence for an entire
period of 12 months.
When collecting data concerning injuries in football, it is recommended that not only the type and
number of injuries but also the match and training
exposure should be reported (Fuller et al., 2006). By
doing so, it is possible to calculate the incidence rates
for training and match play separately. The incidence
rates should be reported as the number of injuries per
1000 exposure hours. It is important that the injury
definition is the same when comparing results from
different studies. We have therefore chosen to use the
definition recommended by the FIFA Injury Consensus Group (Fuller et al., 2006). In the definition
used, all injuries that occurred suddenly to the region
of the posterior thigh sustained during a football
match or training, irrespective of the need for medical attention or time loss from football activities,
were registered as injuries. Injuries with direct
trauma to the posterior thigh (contusion) were excluded. A limitation to this injury definition is that
not all analyzed injuries may have been hamstring
injuries. In other studies, imaging modalities such as
primarily magnetic resonance imaging (MRI) have
Table 3. Hamstring injuries by playing position and age
Playing position
No.
%
Distribution
for the entire
cohort %
Age
distribution
No.
Goalkeeper
Defender
Midfielder
Forward
Alternating
Not specified
Total
0
20
23
7
3
1
54
0
37
42
13
6
2
100
10
28
35
18
8
1
100
17–22
23–28
29–34
351
Not specified
20
24
7
2
1
37
44
13
4
2
44
36
17
2
1
Total
54
100
100
590
%
Distribution
for the entire
cohort %
Acute hamstring injuries in Danish elite football
20
Not specified
18
Match play
Training
No. of hamstring injuries
16
14
12
10
8
6
4
2
Fig. 1. Month of hamstring injuries during training or match.
0
Mar
April
May
been used in order to verify hamstring injuries
(Verrall et al., 2003). However, restricting the diagnosis only to radiologically verified injuries does not
solve this problem. It is known that a huge number of
‘‘characteristic’’ hamstring injuries are not detectable
in MRI or ultrasound (US) (Verrall et al., 2003;
Connell et al., 2004), and there are discrepancies
between findings when comparing MRI and US
(Connell et al., 2004).
One of the aims of this study was to report the
recurrence rate for athletes with a hamstring injury in
the 12-month period before the study. To identify
players with a hamstring injury in the previous 12
months, we had to rely on the reported baseline
information given by the players. Because of the
probability of recall bias, players in the present study
were only asked for any hamstring injuries 12 months
before this study. It has been shown previously that
80% of amateur football players (Australian Rules
football) are able to recall the body region and
number of injuries sustained in the past 12 months
(Gabbe et al., 2003). In the present study, we have
studied elite players and specified the question to
details of hamstring injuries only during the last 12
months. We therefore believe that the chance of recall bias is reduced. The 12% was in accordance with
the studies of Arnason et al. (1996) and Woods et al.
(2004).
In this study, we have defined re-injuries as injuries
of the same type and in the same anatomical site that
recur in the following year after the index injury.
Other studies have used different re-injury definitions
and the recurrence rate is calculated for instance as
the sums of re-injuries within the same and the
previous playing season (Arnason et al., 1996) or
re-injuries within 2 months of the final rehabilitation
June
July
Aug
Sep
Oct
Nov
Dec
Jan
Feb
day of the previous injury (Waldén et al., 2005). We
have chosen to report the recurrence rate for the
following year in order to cover the risk of recurrence
over time. It is well known that setbacks are common
in the rehabilitation period following a hamstring
injury and that players are susceptible to re-injuries
in the first matches played after the rehabilitation
period. In a study by Orchard and Best (2002) from
the Australian Football League, the timing of hamstring re-injuries in the same playing season as the
index injury was reported. The study showed that
more than one-third of the inter-season re-injuries
occurred in the first week after the final rehabilitation
day of the previous injury. These early re-injuries
might be influenced by the management of the
rehabilitation in the clubs, the approach to return
to play after an injury, the player’s eagerness to
return to play, etc. Most of these issues can be
avoided when reporting the recurrence rate in the
following year. There are strengths and weaknesses
in the different ways of defining re-injuries and
reporting recurrence rates, and one should be aware
of the different items covered when choosing which
method to use.
In Fig. 1, the number of hamstring injuries in
training and match play throughout each month
during the 12-month period is shown. The figure
shows an accumulation of injuries in the start of the
second phase of the playing season after the 3.5month mid-seasonal winter break. In the winter
break, all of the participating teams in this study
performed condition and strength training on an
organized level and played practice matches weekly.
All teams had a vacation for 2–3 weeks.
This study has registered hamstring injuries in a
12-month period covering one competitive season
591
Petersen et al.
only. It is therefore not known whether the seasonal
distribution of hamstring injuries seen in this study is
in accordance with the general seasonal distribution.
The seasonal distribution in the present study is
contradictory to the findings of Woods et al. (2004),
who did not find any significant differences throughout each month of the season. One reason for the
findings in the present study could be a decreasing
compliance of reporting injuries throughout the study
period. Another explanation for the different seasonal
distribution seen could be that the condition and
strength training in the 3.5-month mid-seasonal break
did not focus appropriately on the hamstring muscles.
One might speculate that a normal training session
does not adequately reproduce match-playing conditions. This could be caused by different work intensities, running lengths, level of competitiveness, etc.
activity compared with training. Despite the high
injury incidence, evidence of efficacy prevention and
rehabilitation is not well established (Bahr & Holme,
2003; Petersen & Hölmich, 2005; Goldman & Jones,
2007; Mason et al., 2007). Because the injuries are
common in a widespread sport such as football, there
is a need for further research concerning injury
prevention, preferably in the form of randomizedcontrolled trials. This study provides valuable information in the process of study design and power
calculation when planning a randomized-controlled
trial with the aim of preventing hamstring injuries.
Key words: hamstring, muscle, strain, injury, incidence, recurrence, football.
Acknowledgements
Perspectives
This study shows that first-time and recurrent hamstring injuries are common in Danish elite football.
Each team sustained a mean of 3.4 hamstring injuries
per season, with a mean of 21.5 days missed per
injury. This huge amount of time loss due to a single
injury type might be critical for most teams. Furthermore, the study highlights match play as a high risk
We would like to thank the physiotherapists in the participating clubs for their time and trouble during the 1-year period of
injury registration. Thanks are also to all participating players
for their cooperation. We are very grateful for the statistical
supervision of professor Per Kragh Andersen, Department of
Biostatistics, Institute of Public Health, University of Copenhagen, Denmark.
This study was in part supported by a grant from Team
Copenhagen.
References
Arnason A, Gudmundsson A, Dahl HA,
Jóhannsson E. Soccer injuries in Iceland.
Scand J Med Sci Sports 1996: 6: 40–45.
Arnason A, Sigurdsson SB, Gudmundson
A, Holme I, Engebretsen L, Bahr R.
Risk factors for injuries in football. Am
J Sports Med 2004: 32: 5S–16S.
Bahr R, Holme I. Risk factors for sports
injuries – a methodological approach.
Br J Sports Med 2003: 37: 384–392.
Connell DA, Schneider-Kolsky ME,
Hoving JL, Malara F, Buchbinder R,
Koulouris G, Burke F, Bass C.
Longitudinal study comparing
sonographic and MRI assessments of
acute and healing hamstring injuries.
Am J Roentgenol 2004: 183: 975–984.
Ekstrand J, Gillquist J. Soccer injuries
and their mechanisms: a prospective
study. Med Sci Sports Exerc 1983: 15:
267–270.
Ekstrand J, Timpka T, Hägglund M. Risk
of injury in elite football played on
artificial turf versus natural grass: a
prospective two-cohort study. Br J
Sports Med 2006: 40: 975–980.
Fuller CW, Ekstrand J, Junge A,
Andersen TE, Bahr R, Dvorak J,
Hägglund M, McCrory P, Meeuwisse
WH. Consensus statement on injury
definitions and data collection
592
procedures in studies of football
(soccer) injuries. Scand J Med Sci
Sports 2006: 16: 83–92.
Gabbe BJ, Finch CF, Bennell KL,
Wajswelner H. How valid is a self
reported 12 month sports injury history?
Br J Sports Med 2003: 37: 545–547.
Goldman EF, Jones DE. Interventions
for preventing hamstring injuries.
(Protocol) Cochrane Database of
Systematic Reviews 2007, Issue 4. Art.
No.: CD006782. doi: 10.1002/
14651858.CD006782.
Hägglund M, Waldén M, Ekstrand J.
Previous injury as a risk factor
for injury in elite football: a prospective
study over two consecutive seasons.
Br J Sports Med 2006: 40: 767–727.
Hawkins RD, Hulse MA, Wilkinson C,
Hodson A, Gibson M. The association
football medical research programme: an
audit of injuries in professional football. Br J Sports Med 2001: 35: 43–47.
Mason DL, Dickens V, Vail A.
Rehabilitation for hamstring injuries.
Cochrane Database of Systematic
Reviews 2007, Issue 1. Art. No.:
CD004575. doi: 10.1002/
14651858.CD004575.pub2.
Orchard J, Best TM. The management of
muscle strain injuries: an early return
versus the risk of recurrence. Clin J
Sports Med 2002: 12: 3–5.
Orchard J, Best TM, Verrall GM.
Return to play following muscle
strains. Clin J Sports Med 2005: 15:
436–441.
Orchard J, Seward H. Epidemiology of
injuries in the Australian Football
League, season 1997–2000. Br J Sports
Med 2002: 36: 39–44.
Petersen J, Hölmich P. Evidence based
prevention of hamstring injuries in sport.
Br J Sports Med 2005: 39: 319–323.
Verrall GM, Slavotinek JP, Barnes PG,
Fon GT. Diagnostic and prognostic
value of clinical findings in 83 athletes
with posterior thigh injury. Am J
Sports Med 2003: 31: 969–973.
Waldén M, Hägglund M, Ekstrand J.
UEFA Champions League study: a
prospective study of injuries in
professional football during the 2001–
2002 season. Br J Sports Med 2005: 39:
542–546.
Woods C, Hawkins RD, Maltby S, Hulse
M, Thomas A, Hodson A. The football
association medical research
programme: an audit of injuries in
professional football – analysis of
hamstring injuries. Br J Sports Med
2004: 38: 36–41.