Article Text

Download PDFPDF

Fracture epidemiology in male elite football players from 2001 to 2013: ‘How long will this fracture keep me out?’
  1. David Larsson1,2,
  2. Jan Ekstrand2,3,
  3. Magnus K Karlsson1
  1. 1Department of Clinical Sciences and Orthopedic Surgery, Skåne University Hospital, Lund University Malmö, Malmö, Skåne, Sweden
  2. 2Football Research Group, Linköping University, Linköping, Sweden
  3. 3Division of Community Medicine, Department of Medical and Health Sciences, Linköping University, Linköping, Sweden
  1. Correspondence to David Larsson, Department of Clinical Sciences and Orthopedic Surgery, Skåne University Hospital, Lund University Malmö, Riksdalervägen 9, Hägersten S-129 32, Sweden; david.mikael.larsson{at}


Background Determining fracture risk and rehabilitation periods after specific fractures in professional football is essential for team planning.

Aim To identify fracture epidemiology and absences after different types of fractures in male professional football players.

Methods 2439 players from 41 professional male teams in 10 countries were followed prospectively from 2001 to 2013. Team medical staff registered fractures, absences after fractures and player exposure.

Results 364 fractures were recorded, with an incidence of 0.27/1000 h of exposure (95% CI 0.25 to 0.30). The incidence of traumatic fractures was 0.25 (0.22 to 0.27) and that of stress fractures was 0.03 (0.02 to 0.04). 45% of traumatic fractures and 86% of stress fractures affected the lower extremities. Absence after a fracture was 32 days (1–278) (median (range)), compared to that after a traumatic fracture of 30 days (1–278) and a stress fracture of 65 days (6–168) (p<0.001). Annual fracture incidence was stable during the study period (R2=0.051, b=−0.011 (95% CI −0.043 to 0.021)). Young players had a relative risk of 10.9 (3.3 to 35.6) of sustaining stress fractures compared to old players (p<0.01). The fracture incidence did not differ between individuals in different playing positions (p=0.10).

Summary A male professional football team can expect 1 to 2 fractures per season. There are more traumatic fractures than stress fractures; while most fractures affect the lower extremities, stress fractures yield longer absences than traumatic fractures and young players have more stress fractures than old players. There is no difference in risk among players at different playing positions.

  • Epidemiology
  • Soccer
  • Injury

Statistics from

Request Permissions

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


Fractures constitute 3% of all injuries in male professional football, corresponding to 1–2 fractures per team and season,1 while constituting 12% of all injuries among amateur players.2 Stress fractures in elite male football players are reported to have an incidence of 0.04/1000 h of exposure,3 while the incidence of traumatic fractures among professionals is unclear. The general absence from sports after a stress fracture is 3–5 months,3 and after a traumatic fracture 2–9.5 months depending on the fracture location.4–7 It is therefore essential for the planning, both for the individual and the team, to identify specific rehabilitation periods after each type of fracture.

Amateur players reportedly sustain most traumatic fractures in the upper extremities,8 ,9 while most stress fractures affect the lower extremities, particularly in the fifth metatarsal bone.3 It is also well established that match play confers more injuries than training sessions.1 ,2 ,10–20 However, it remains to be determined if fracture distribution and type of exposure when sustaining a fracture are similar in professional players. Young age is another risk factor for injuries and stress fractures in male football players.3 ,21 ,22 It has also been shown that midfielders cover the most distance during matches,23–25 but whether players in this position suffer the most stress fractures is unknown. Defenders and forwards are exposed to most player-to-player contact leading to injuries,26 ,27 but it is not established if players in these positions suffer the most traumatic fractures.

We hypothesised that among male professional football players, (1) fractures are rare compared to other types of injuries, (2) fractures occur more often during match play than during training, (3) traumatic fractures are more common than stress fractures, (4) most traumatic fractures occur in the upper extremities while most stress fractures occur in the lower extremities, (5) stress fractures are associated with a longer recovery than traumatic fractures, (6) the fracture incidence has been stable during the past decade, (7) young age is a risk factor for fractures, (8) playing positions with many contact situations are risk factors for traumatic fractures and (9) playing position with long running distance is a risk factor for stress fractures.

Materials and methods

Study procedure

This study was conducted using data from the ongoing UEFA Elite Club Injury Study; the study design is described in detail in previous reports.28 In summary, this study collects ongoing data from a number of professional football clubs in Europe since 2001. Baseline data are collected at the start of each season, containing information such as age, weight, height and playing position. The minutes that the players participate in matches and training sessions (defined as player exposure) are prospectively recorded and continuously sent to the UEFA Champions League (UCL) Elite Club Injury Study group by the medical staff. Medical staff registered all injuries on a specific injury form that identified the type of injury, when the injury was sustained and location of the injury. The medical team also classified the injury as being traumatic or an overuse injury, if the injury was a first-time injury or a re-injury and the recovery period before the players can return to football. The definitions (table 1) and the procedures of data collection were all according to the UEFA guidelines.28 All clinical symptoms of a fracture, no matter the subtype, were therefore verified by imaging before being reported as a fracture. To estimate the prevalence of fractures among all types of injuries, we also registered non-fracture injuries. The UEFA Medical Committee and the UEFA Football Development Division have approved the study design.

Table 1

Operational definitions used in the study

Study population

In this report, we followed 41 male teams from the top divisions of 10 countries in Europe prospectively during 1–12 seasons (total 204 team seasons) from June 2001 to end of the season 2012/2013. The final database included 2439 players aged 25.6±4.6 years (mean±SD) at inclusion who were then followed for 1 337 094 h of football exposure (1 123 623 h of training and 213 470 of match play). We defined players below the mean age of all players (25.6±4.6 years) as young players (aged 22.1±2.1 years) and those above as old players (aged 29.6±2.9).


We used IBM SPSS Statistics V.22 for the statistical analyses. Data are presented as numbers and proportions (%), mean±SDs, median with range or mean with 95% CIs. Fractures were expressed as numbers and incidence as the number of sustained fractures per 1000 h of match play or football training. Group comparisons were done using χ2 test for categorical variables, Mann-Whitney U test or for non-normally distributed continuous variables and analysis of variance for multiple group comparisons. Seasonal trends were analysed using linear regression with the log transformed fracture incidence as the dependent variable. A p value <0.05 was regarded as a statistically significant difference.


Fracture epidemiology

There were 364 fractures (3.5% of all injuries) and 9891 other injuries during the study period (table 2). Refractures constituted 8% of all fractures (table 2). The fracture incidence was 0.27 (0.25 to 0.30) per 1000 h of exposure, with an incidence during training of 0.10 (0.09 to 0.12) and during match play of 1.17 (1.03 to 1.32). The incidence of traumatic fractures was 0.25 (0.22 to 0.27), during training 0.09 (0.07 to 0.11) and during match 1.08 (0.95 to 1.23). The incidence of stress fractures was 0.03 (0.02 to 0.04). Thus, the fracture incidence was 12 times higher during match play than during training (p<0.001). The incidence of traumatic fractures was eight times higher than the incidence of stress fractures (p<0.001).

Table 2

Number of fractures, proportion of all fractures (%), incidence of fractures (per 1000 h of exposure to football) (mean (95% CI), absence from football in days after a fracture (mean±SD and median (IQR) number of refractures and proportion of refractures among all fractures (%) in male elite football players

Fracture location

The location of the most common fractures, the incidence and the proportion of all fractures are shown in table 2. In total, 43% of all fractures occurred in the ankle, feet or toe while 12% occurred in the wrist, hand or fingers and 8% of all fractures were refractures. In total, 45% of the traumatic fractures and 86% of the stress fractures occurred in the lower extremities. Only 23% of the traumatic fractures and none of the stress fractures occurred in the upper extremities.

Fractures and recovery period

The median absence after a fracture was 32 days (range 1–278). The median absence after a traumatic fracture was 30 days (range 1–278), and after a stress fracture it was 65 days (range 6–168) (p<0.001). The median absence after any type of primary fracture was 30 days (range 1–278), and after a refracture it was 65 days (3–155) (p=0.005). The median absence from football after fractures in the head and neck regions was 11 days (1–93), in the trunk region 21 days (range 2–104), in the upper extremities 24 days (1–144) and in the lower extremities 69 days (2–278) (p<0.001) (table 2).

Fractures and time trends

There was no change in annual fracture incidence over the studied seasons (R2=0.051, b=−0.011, 95% CI −0.043 to 0.021) (figure 1). There were no further changes in traumatic fracture incidence (R2=0.121 b=−0.016, 95% CI −0.047 to 0.015) or in stress fracture incidence (R2=0.060, b=0.033, 95% CI −0.074 to 0.141) during the same period (figure 2).

Figure 1

Incidence of fractures in male elite football players each season. Incidence is expressed as the number of fractures per 1000 hours of exposure to football. Data is presented as total incidence with 95% confidence intervals, represented by error bars.

Figure 2

Incidence in male elite football players of traumatic and stress fractures each season. Incidence is expressed as the number of traumatic and stress fractures per 1000 hours of exposure to football. Data is presented as total incidence with 95% confidence intervals, represented by error bars.

Fractures and age

The mean age when sustaining a fracture was 25.2±4.4 years (mean±SD). Young players had a fracture incidence of 0.32 (0.28 to 0.36) per 1000 h of exposure, compared to 0.23 (0.19 to 0.27) in old players, resulting in a rate ratio of 1.40 (1.13 to 1.72).

The traumatic fractures incidence among young players was 0.27 (0.23 to 0.31) compared to 0.22 (0.19 to 0.26) in old players, resulting in a non-significant rate ratio of 1.20 (0.97 to 1.50).

The stress fracture incidence among young players was 0.05 (0.03 to 0.07) compared to 0.22 0.005 (0.002 to 0.014)) in old players, resulting in a rate ratio of 10.9 (3.3 to 35.6). Stress fracture incidence was 11 times higher in young players than old players (p<0.001).

Fractures and playing position

There was no statistically significant difference in fracture incidence (p=0.10), traumatic fracture incidence (p=0.11) or stress fracture incidence (p=0.99) when comparing players at different positions (table 3).

Table 3

Incidence of fractures with 95% CIs divided for goalkeepers, defenders, midfielders and forwards, subdivided for match, training and subdivided for traumatic and stress fractures


This study shows that in male professional football players, the incidence of fractures is higher among young players than old players. Rehabilitation periods following fractures are in general long and generally longer for stress fractures than traumatic fractures, but vary in duration depending on anatomic location. Furthermore, fractures are rare compared to other type of injuries and occur more often during match play than during training. A team of 25 players, being exposed to around 5600 h of football per year, can with the incidences we found expect 1 to 2 fractures per season, and this incidence has been stable during the past decade. Both traumatic and stress fractures primarily affect the lower extremities and playing position is not a risk factor.

Stress fractures are less common than traumatic fractures but cause longer absences

In this study, traumatic fractures were almost 10 times as common as stress fractures. However, stress fractures, even if uncommon, comprise a clinical and economic problem since the mean absence from football is longer after a stress fracture than after a traumatic fracture. The reason could be that stress fractures affect the load-bearing lower limbs more often than traumatic fractures. While fractures in unloaded regions could allow the player to play football even with an unhealed fracture, and even though absences after fractures to unloaded regions are shorter than after fractures to the lower extremities, it is noteworthy that, for example, forearm fractures are associated with a high risk of refractures (table 2). It could therefore be questioned if rehabilitation periods should not be extended after these types of fractures.

Fractures among professionals affect the lower extremities and differ in time lost

This study extends previous publications that report fractures as uncommon compared with other injuries in sportsmen.1 ,2 ,8 The proportion of fractures (3.5% of all injuries) was one-quarter of what has been reported in amateur players.2 Most fractures in professional male football players occurred in the lower extremities, in contrast with reports in amateur players, where most fractures have been inferred to occur in the upper extremities.8 ,9 We speculate that these discrepancies may relate to the more intense play, different training methods, different bone mass and neuromuscular function.

The literature also infers that the mean absence from football following a traumatic fracture is around 3.5 months in the general population.8 This is twice as long as in our study (table 2). We also note that several fractures had a shorter rehabilitation period than has been reported in amateur sportsmen. For example, the mean absence after a fibula fracture was in our study 11 weeks compared to 18 weeks in other professionals and non-professional sportsmen,6 and 12.5 weeks after an ankle fracture compared to 17–26 weeks in other sportsmen.29 ,30

We speculate that since the teams in our study have the ideal access to supervision, medical expertise, physical therapy, mental coaching and financial resources to speed recovery, our data may represent the ideal healing and recovery. For example, ankle rehabilitation programmes with physiotherapy and exercise after an ankle fracture have been shown to facilitate rehabilitation in the general population.31 Perhaps this level of care has a part in the explanation for the short rehabilitation times among professionals.

In our study, the median absence following a facial fracture was 19 days, with a low refracture risk (table 2), whereas the recommended rehabilitation time in other sports is 6–8 weeks.30 ,32 ,33 It should be noted, however, that our data on facial fractures exclude mandible fractures, as these constitute a separate category on team injury report sheets, possibly explaining the shorter rehabilitation times following facial fractures in our study. Another possible explanation is the use of facial protective masks following a facial fracture among professional players.34 However, since we lack data on the management of facial fractures, this remains a speculation.

An ideal supervision in the rehabilitation process may also decrease the risk for refractures, a most important issue since refractures had longer recovery periods than primary fractures. An interesting question then arises; if similar medical and rehabilitation conditions would be available in the general population, could this decrease the sick leave and facilitate the return to work? Future studies to evaluate this could be advantageous.

Young players suffer more stress fractures

Young players suffered more fractures than old players, due to a higher incidence of stress fractures among young players. Our data support the literature in which stress fractures, especially in the fifth metatarsal bone, has been linked to young age.3 ,21 One explanation could be that individuals whose anatomy is insufficient for elite football achieve their injuries at a young age and then stop their elite careers. Thus, older players could represent a selection of those who are capable of playing football on a high level without sustaining stress fractures. Another explanation could be that older players' musculoskeletal structure has been accustomed to a training load better than the younger players, due to the longer training history, and thus having less overload injuries. Yet another explanation could be that older players are more prone to sustain other overuse injuries, as the incidence of certain muscle strains is shown to increase with player age,11 and then reduces their activity level before having a stress fracture. Since we can only speculate in regard to causality, future research ought to clarify the reasons behind the reported age effects on fracture preponderance in male elite football players.

Playing position does not influence fracture incidence

Our study opposes the hypothesis that defenders and forwards have more traumatic fractures than other players and that midfielders have more stress fractures. Our inferences are supported in the literature,35–39 making us draw the conclusion that distance covered during matches and tackling patterns23 ,40 cannot explain any differences in fracture epidemiology in elite football players. Our results are in line with other studies, where playing position has not been shown to influence injury incidence,35–39 but contradict others.16 ,26 ,27 ,40–43 Methodological discrepancies among the studies, as pointed out by Kristenson et al,16 could at least partly explain the differences.

Methodological considerations

The strengths of this study include the homogeneous cohort, the large sample size, the prospective study design, the inclusion of only objective verified fractures, the inclusion of different types of fractures and the separate evaluation of different fracture types and the rehabilitation period for each specific fracture. Our data could therefore be used by medical personnel to estimate rehabilitation times. The limitations include the lack of data on the management of injuries as well as female and amateur football data.


This study can conclude that male professional football teams experience 1–2 fractures per year; where this incidence has been stable from 2001 to 2013, fractures occur more commonly during match play than during training; traumatic fractures are more common than stress fractures, and most fractures occur in the lower extremities; young players have more fractures than old players, there is no different fracture risk in players at different playing positions and stress fractures render a longer rehabilitation period than traumatic fractures. Future studies are recommended to search for causality that could explain the higher incidence in young players and if intervention strategies could reduce the incidences and rehabilitation periods.

What are the findings?

  • Fractures represent 3.5% of all injuries in male professional football, with an incidence of 0.27 fractures per 1000 h of exposure.

  • Most fractures affect the lower extremities.

  • Stress fractures are less common than traumatic fractures.

  • Stress fractures cause longer absences from football than traumatic fractures.

  • The fracture incidence has been stable during the past decade.

  • Young age is a risk factor for fractures.

  • Playing position is not a risk factor for fractures.

  • Rehabilitation times following fractures seem to be shorter in male elite football players than after similar fractures in non-professional sportsman and the general population.

  • Rehabilitation times for stress fractures are longer than after traumatic fractures.

How might it impact on clinical practice in the future?

  • The team medical and coaching staff should be aware of the increased risk for young players to suffer stress fractures, as these generally require long rehabilitation times.

  • The reported rehabilitation times for specific fractures could serve as a base of knowledge and recommendation for team medical staff.


The authors would like to thank all the participating clubs, including coaches, players and medical staff. They also thank Dr Mark Waller for language revisions.


View Abstract


  • Contributors DL, JE and MKK were responsible for the conception and design of the study. All authors contributed to interpretation of the findings. DL wrote the first draft of the paper, which was critically revised by JE and MKK. The final manuscript was approved by all authors.

  • Funding The Football Research Group has been established in Linköping, Sweden, in collaboration with Linköping University and through grants from the UEFA, the Swedish Football Association, the Football Association Premier League Limited and the Swedish National Centre for Research in Sports.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval The study design was approved by the UEFA Medical Committee and the UEFA Football Development Division.

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