Article Text

Download PDFPDF

Men at higher risk of groin injuries in elite team sports: a systematic review
  1. John William Orchard
  1. Correspondence to Dr John William Orchard, School of Public Health, Western Ave, University of Sydney 2006, Australia; johnworchard{at}


Background Groin injuries are common in sports, particularly multidirectional team sports, but incidence rates across sports other than football (soccer) have been poorly documented.

Methods A systematic review (initially using PubMed and SportDiscus databases) was performed to record incidence of groin and groin region injuries in sports. Inclusion criteria included presentation of groin injury incidence data for at least 10 team/squad seasons.

Results Data from 31 studies were included. These used varying injury definitions and also considered varying injury categories from general to specific (all groin/hip region injuries, groin injuries, adductor muscle strains, intra-articular hip injuries). When playing the same sport, men had greater injury incidence of groin injury than women (relative risk, RR 2.45, 95% CI 2.06 to 2.92). Sports with high incidences of groin injury included ice hockey and the football codes. There is variation by player position for rate of groin injury in many sports. Hip injuries have become more commonly diagnosed over the past decade in Australian football (p=0.001) and other sports.

Conclusions There is moderate evidence that men have a higher risk of groin injury than women when playing the same sport. There is some evidence that hip injuries are being increasingly diagnosed in the subset of ‘groin injuries’ in recent years. It is recommended that injury epidemiology consensus statements aim to include a number of relevant sports to improve injury incidence comparisons among different sports.

  • Groin
  • Epidemiology
  • Surveillance
  • Hip

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.


Groin injuries are especially common in elite level multidirectional team sports. However, the exact epidemiology of groin injuries is poorly documented. This is partly due to diagnostic difficulty.1 Although thorough studies of interclinician reliability for groin injury diagnoses have not been formally published, it is well recognised that there is potentially widespread variability in diagnosis between clinicians.2–6

Diagnostic uncertainty means that, to date, epidemiological studies in sports medicine have tended to group most musculoskeletal causes of groin pain together in a combined category of ‘groin injuries’, which can include acute adductor muscle strains as well as other common causes of chronic groin pain. Some injury surveillance systems use the entire ‘groin/hip/pelvis’ as a body region category, whereas others differentiate ‘hip joint injuries’ (which includes labral tears and femoroacetubalar impingement) from ‘groin injuries’.

Comparisons of the rates of groin injuries between sports are lacking. This review aimed to document incidence of groin injuries published arising from multiyear (and/or multiteam/squad) sports injury surveillance systems in a variety of sports. The most comprehensive data published regarding incidence rates of groin injuries are on (soccer) football.4 ,7 ,8 In this paper the term ‘soccer’ is used to differentiate it from the other football codes (American football, Australian football, Gaelic football, rugby league and rugby union).


Search strategy and inclusion criteria

The PRISMA statement was consulted when designing strategy for this review,9 although no intervention was being assessed which is the underlying basis for PRISMA.

The Pubmed and Sport Discus databases were used to assist with searching for papers which detailed sports groin injury incidence. The following initial search strategy was used

(Football OR rugby OR athletics OR swimming OR cricket OR baseball OR hockey) AND (incidence OR prevalence OR epidemiology) AND (groin OR hip OR adductor) AND injury.

The following criteria were used to exclude papers from the review

  1. If groin incidence data (ie, including exposure data) were not presented.

  2. Fewer than 10 team/squad seasons of data were presented (ie, <10 seasons for an individual team/squads; at least 10 teams in a single year survey; at least five teams for a 2 year survey) to exclude studies with small sample size that may not provide meaningful comparative data.

  3. Involving only the sport of soccer football (although studies were included if soccer incidence data were presented in comparison with other sports using a similar methodology).

  4. Publication superseded by subsequent publication of same cohort (although again included if a direct comparison to other sports involved).

Statistical analysis

To enable comparison with as many other studies as possible, published incidence rates were reproduced or converted to one or more of the following units

  1. Injury incidence per 1000 player hours. An exception was a group of US college sport papers where incidences are presented in the related original unit of injuries per 1000 athlete exposures.

  2. Injuries per 100 player seasons (ie, where injuries were presented as number per team per season, this unit was converted to 100 based on team size in the definition).

  3. Groin (or groin region) injuries presented as a percentage of all injuries.

For sports where comparable male and female injury incidence was available, a RR comparison was made with 95% CIs.10 When the 95% CIs for the RR did not contain the value 1.0, this was taken to indicate a statistically significant difference between sexes. A pooled RR was performed between male and female injury rates in all available sports where comparative data were available. Potential confounders were not assessed statistically in this paper.

For studies where multiyear (≥5 year) statistics were presented with annual data, the LINEST function in Excel was used to determine trends over time to see whether groin injuries (or related diagnoses) were increasing or decreasing over time.

Levels of evidence

When there was sufficient methodological homogeneity to justify comparing the results between studies, qualitative analysis of the data was carried out using a levels of evidence system.

  1. Strong evidence: provided by two or more studies with high quality and by generally consistent findings in all studies (≥75% of the studies reported consistent findings).

  2. Moderate evidence: provided by one study with high quality and/or two or more studies with low quality and by generally consistent findings in all studies (≥75% of the studies reported consistent findings).

  3. Limited evidence: provided by only one study with low quality.

  4. Conflicting evidence: inconsistent findings in multiple studies (<75% of the studies reported consistent findings).

  5. No evidence: when no studies could be found.


Search results

The SPORTDiscus (116 results) and Pubmed (486 results) databases were searched on 4 September 2014 according to the search strategy. After addition of records identified through other sources, removal of duplicates and screening of firstly titles and then abstracts for relevance, 48 articles were retrieved in full text. A further 17 studies were then excluded as insufficient groin incidence data were presented. This resulted in 31 studies used for analysis in this review. Of these, 15 papers were all published in a companion series comparing injury incidence in the USA major NCAA (National Collegiate Athletic Association) sports.

Comparative injury incidence in US College Sports

A large series of studies were presented in a single special issue of the Journal of Athletic Training which compared injury incidence from a multitude of sports arising from the NCAA injury database. All of these papers, summarised in table 1, included injury incidence data for the category ‘Pelvis/hip’ (region) ‘Muscle/tendon’ (category) injuries. Incidence data were presented as an incidence rate per 1000 athlete exposures for games and practices (training sessions). Incidence was also expressed a percentage of all injuries that were groin/hip muscle/tendon injuries. The sports with the highest rates of groin injuries were men's and women's ice hockey and men's soccer (table 1).

Table 1

Comparison of US collegiate pelvis/hip muscle/tendon injury rates for 15 sports

For three sports (soccer, ice hockey and basketball) comparative data were available for men and women (table 2). This showed that men had higher groin injury rates than women in all three sports, which was statistically highly significant in two of them. The combined RR for all three sports also found a significantly increased risk for men.

Table 2

Comparison of US collegiate male and female groin injury rates for three sports

Comparative incidence in elite male team sports

The other studies included all related to elite male team sports. This reflects the reality that individual and amateur sports have not generally reached the point of multisquad multiyear injury surveillance systems which can document specific diagnoses. Even elite level individual sports like athletics and swimming are not yet implementing and publishing results of long-term category-specific injury surveillance systems. As the NCAA studies contained female sport injury data, male and female sports remained in this review, but the final paper title differs from the initial search strategy to reflect that the review only revealed long-term groin region incidence data from elite team sports (table 3).

Table 3

Incidence of groin, hip and abdominal injuries in various elite male team sports

Table 3 also reveals great variation in the methodology of reporting of groin region injury incidence. Some studies used an injury definition which included all medical presentations, whereas others screened medical presentations to only include injuries which led to either missed training or match time, or more narrowly simply missed match time. Furthermore, some studies combined injuries which had a match or training onset, whereas in others match injuries and training injuries were separated. Some studies reported incidence of a specific injury diagnosis (eg, adductor muscle strains) whereas others reported incidence for broader categories (groin injuries) or even region only (all hip and groin region injuries, which includes contact mechanism injuries such as haematomas).

There was only one study which compared groin injury rates in different sports with the same methodology. This study found that Australian football had a higher rate of groin injuries than rugby league and rugby union.26

Variation in groin injury incidence by position within sport

In addition to the variation in rates between sports, multiple studies have reported variation in groin injury rates within sports by player position (table 4). In rugby union, backs (who do more running and kicking) have higher rates of groin injury than forwards (who do more tackling and other body contact).38 In baseball, pitchers also had a higher rate of groin injuries and total injuries than position players (fielders/batters). 37 In ice hockey, goaltenders had a higher rate of intra-articular hip joint pathology than forwards or defensemen, although all positions had high rates of groin injuries.27 In cricket, fast bowlers have higher rates of most injuries than players in other positions, including groin injuries.29 In American footballers, Special Team players (who do more change-of-direction running and kicking) have a higher rate of hip injuries compared to offensive and defensive players.28

Table 4

Comparison of groin injury incidence by position in various sports

Injury trends over time in the various sports

The only study which reported annual rates of groin region and hip joint injuries over a long time period examined annual incidence from the Australian Football League (AFL).41 Table 5 reveals that groin injuries and groin region injuries have not changed significantly over time, but that incidence of hip joint injuries has increased annually by +0.11 (additional hip joint injuries per team per season) over the last decade (p=0.001).

Table 5

Incidence (injuries per 40 man squad per season) of groin and hip injuries in the Australian Football League


This review of 31 epidemiological studies on groin injuries found a wide array of injury definitions and categories making comparisons hard. In general men had a higher incidence than women when playing the same sport (RR 2.45, 95% CI 2.06 to 2.92). Sports with high incidences of groin injury included ice hockey and the football codes. There is variation by player position for rate of groin injury in many sports. Hip injuries have become more commonly diagnosed over the past decade in Australian football (p=0.001) and other sports.

Variation noted in reported incidence measures and study limitations

In sports injury epidemiology, some sports have written consensus statements for reporting guidelines in these sports, such as cricket, soccer and rugby union.42–44 There is no common guideline used across multiple sports which can be used to compare and combine data. In the absence of guidelines, the methods of this review decided on cut-offs for inclusion of studies (10 team seasons worth of data; incidence rates reported). The heterogeneity of incidence reporting of the studies (particularly the definition of what constituted a ‘groin’ injury) has meant that direct comparison of rates among the majority of included studies was not possible. Although individual sports have published consensus statements,42–44 with varying levels of uptake, to try to enable comparison of studies within sports, there would certainly be value in trying to develop a multisport consensus statement, if possible.

In planning this review, attention was given to some aspects of the PRISMA statement.9 The STROBE statement45 which advises on reporting of observational epidemiological data were also consulted. STROBE is not focused on meta-analysis or combining of multiple studies. This means that at the current time there is no accepted methodology for a systematic review of sports injury epidemiology. A more specific set of guidelines could be developed to address this. An alternative would be development of advice on how to incorporate the relevant aspects of the existing PRIMSA and STROBE methodologies.

Variations between studies exhibited in this review could be used as a basis to try to achieve consensus. To date consensus studies have tended to focus on injury definitions (ie, whether to include ‘all’ injuries, medical presentation injuries, training-time and match- time-loss injuries or just match-time-loss injuries). However, the demarcation of injury categories is equally important, as it is difficult to compare, say, 5 ‘groin’ strains per 1000 player hours to 3 ‘adductor’ strains per 1000 player hours. The preference for authors to use injury incidence units of either injuries per 1000 player hours or injuries per 1000 athlete exposures also makes comparisons problematic. In some sports, 1 h of play is all ‘in play’ whereas in other sports 1 h of play may contain substantial time where play has stopped and injury exposure is far less. Measuring training time is also very difficult for all sports; if, say a team has an optional weights session in the morning followed by a compulsory field session in the afternoon, does this count as two training exposures per athlete or a single exposure? If an athlete was unable to partake in the optional weights session because of injury but was fit for the compulsory field session, would that qualify as a training-time-loss injury? To account for all of the variations between sports, it may be more practical to measure injury incidence units in ‘injuries per 100 players/season’. This would present a different issue of trying to fairly define what constitutes an ‘average season’ for that sport (ie, how many matches and associated training sessions?). However, it would perhaps be a more realistic way to compare injury incidence between sports, along with being easier to conceptualise than a ‘player hour’.

Men appear to have a higher rate of groin injury than women

This review found evidence that, when playing the same sport, female athletes have lower incidence rates of groin injuries than men. Since this was revealed using three studies with almost identical methodology as part of the NCAA injury surveillance system, it is likely to be a true finding. It is worth noting that the difference in incidence rate between men and women in ice hockey itself did not reach statistical significance but the rate was still higher in men. There is unlikely to be a bias against the reporting of injuries in female athletes in the NCAA injury surveillance system, as this system has previously reported higher rates of ACL injuries in female athletes.46 As there is only one injury surveillance system (NCAA) which has revealed this finding, the level of evidence is ‘moderate’. The exact reasons for men having higher rates of groin injuries than women are not clear from this review but could be the subject of future research.

Ice hockey and football code players are at high risk of groin injury

General observations can be made about the RR of sports based on the data revealed in this review, although conclusions should be more guarded due to the high degree of methodological heterogeneity between studies. This means that it is not possible to state that one particular sport in this review exhibited the ‘highest’ incidence of groin injuries.

Only one study compared multiple sports directly using the same methodology,26 which found that Australian football has a higher rate of groin injuries than rugby league and rugby union. Within the football codes there are higher RRs in certain positions. These positions are the ones where kicking is probably more frequent. This means that the rate of kicking may be correlated to groin injury risk. For example, positions within football codes which kick rarely (offensive and defensive positions in American football and forwards in rugby league and union) appear to have lower rates of groin injuries than other footballers who kick more regularly. Groin injury incidence appears to be high in Gaelic football and Australian football, where the kicking is of the ‘punt’ variety which leads to greater range of hip motion being utilised, combined with regular running and change-of-direction. Other differences between these sports may also play a role, such as pitch size, amount of distance covered and number of changes of direction per player. Future studies should aim at further analysing the movement characteristics of the various sports.

Intra-articular hip joint conditions appear to be increasingly diagnosed

Hip joint conditions (particularly labral tears and femoroacetabular impingement) have been recognised as a possible cause of chronic groin pain in recent years.47 There is a significant increase in hip joint injuries being diagnosed in the Australian Football League over the last decade.41 In addition, the studies from table 3 which detailed rates of hip joint injuries or intra-articular hip pathology have tended to be more recently published than the ones only detailing groin injury rates. The study of hip injuries in the NFL28 showed a trend towards increased diagnosis of hip injuries in recent years, although groin injuries were not presented in the same study. There has also been a documented increase in youth ice hockey hip injuries in Canada in recent years.48 However, there has not been any documented increase in the rate of groin injuries diagnosed in cricket at a time when other injuries have been increasing.29

While it is possible that these trends represent an actual change in the relative occurrence of these injuries (groin vs hip joint) in various sports, it is also possible that the relative change in incidences reflect diagnostic changes (ie, pain which may have been diagnosed as being due to a ‘groin injury’ in the late 1990s is now more likely to be diagnosed as specific hip joint pathology). This is particularly done in cases where hip surgery is undertaken. It is logical that the hip joint will present with pathology in athletes, as the other major joints of the body all develop pathology in elite sport. It is also well-established that retired athletes are more likely to develop hip osteoarthritis.49


This review of 31 epidemiological studies found:

  1. A large degree of heterogeneity with regard to study methodology, injury definitions and terminology and the way in which groin/hip injuries are categorised.

  2. Moderate evidence that men have a higher incidence of groin injuries than women when playing the same sport.

  3. The sport of ice hockey and positions within the football codes which involve a regular combination of kicking and change-of-direction have a high incidence of groin injuries.

  4. Hip joint injuries are being increasingly diagnosed in Australian Football in recent years.

  5. Consensus statements across sports could help address the issues of comparisons between sports with regard to variations of injury definitions, injury categorisation and units of injury incidence.

What are the new findings?

  • A large degree of heterogeneity with regard to study methodology, injury definitions and terminology and the way in which groin/hip injuries are categorised.

  • Moderate evidence that men have a higher incidence of groin injuries than women when playing the same sport.

  • The sport of ice hockey and positions within the football codes which involve a regular combination of kicking and change-of-direction have a high incidence of groin injuries.

How might it impact on clinical practice in the near future?

  • Consensus statements across sports could help address the issues of comparisons between sports with regard to variations of injury definitions, injury categorisation and units of injury incidence.

  • Athletes with groin pain in a given sport who have a choice of position may choose to move to a position which places less relative load on the groin.

  • Attention to hip joint cofactors should be considered in cases of chronic groin pain.



  • Paper submitted to BJSM (as part of special edition) and for presentation at Aspetar 1st World Conference on Groin Pain, Doha, November 2014.

  • Competing interests However, this paper was submitted as part of the 1st World Conference on Groin Pain, for which the author received an honoraruim and a travel grant from Aspetar for attendance.

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