Aim To describe the epidemiology of National Collegiate Athletic Association (NCAA) men's and women's soccer injuries during the 2009/2010–2014/2015 academic years.
Methods This descriptive epidemiology study used NCAA Injury Surveillance Program (NCAA-ISP) data during the 2009/2010–2014/2015 academic years, from 44 men's and 64 women's soccer programmes (104 and 167 team seasons of data, respectively). Non-time-loss injuries were defined as resulting in <24 h lost from sport. Injury counts, percentages and rates were calculated. Injury rate ratios (RRs) and injury proportion ratios (IPRs) with 95% CIs compared rates and distributions by sex.
Results There were 1554 men's soccer and 2271 women's soccer injuries with injury rates of 8.07/1000 athlete exposures (AE) and 8.44/1000AE, respectively. Injury rates for men and women did not differ in competitions (17.53 vs 17.04/1000AE; RR=1.03; 95% CI 0.94 to 1.13) or practices (5.47 vs 5.69/1000AE; RR=0.96; 95% CI 0.88 to 1.05). In total, 47.2% (n=733) of men's soccer injuries and 47.5% (n=1079) of women's were non-time loss. Most injuries occurred to the lower extremity and were diagnosed as sprains. Women had higher concussion rates (0.59 vs 0.34/1000AE; RR=1.76; 95% CI 1.32 to 2.35) than men.
Conclusions Non-time-loss injuries accounted for nearly half of the injuries in men's and women's soccer. Sex differences were found in competition injuries, specifically for concussion. Further study into the incidence, treatment and outcome of non-time-loss injuries may identify a more accurate burden of these injuries.
- Injury prevention
Statistics from Altmetric.com
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.
Soccer is a popular sport worldwide, with strong participation at the US collegiate level.1 In the 2013/2014 season, 23 603 men and 26 358 women participated in National Collegiate Athletic Association (NCAA) soccer.2 Previous research estimates that the injury rate in NCAA men's soccer was 18.75 and 4.34/1000 athlete exposures (AE) in games and practices, respectively; 16.44 and 5.23/1000AE in games and practices, respectively, in NCAA women's soccer.3 ,4 These estimates, however, were limited to time-loss injuries that resulted in at least 1 day lost from sport. While this assists with understanding soccer injuries that limit participation time, it does not address the multitude of injuries that do not result in time loss. Previous research has found substantial proportions of injuries that were not time loss across all sports.5 ,6 Thus, non-time-loss injuries may not currently be well represented in the literature.7
At this time, there is no consensus regarding the definition of a reportable injury for use in US collegiate or high school injury surveillance, and the inclusion of non-time-loss injuries remains a debated issue within sports injury surveillance as a whole.5 ,8–10 The use of multiple injury definitions (including non-time loss) within injury research and surveillance has been proposed; within soccer specifically.11 Fuller et al 11 presented a comprehensive consensus injury definition protocol, to be implemented within soccer injury surveillance, which included non-time-loss and time-loss injuries, categorised as medical attention, and time-loss injuries. The inclusion of non-time-loss injuries may give a truer estimate of injury burden on athletes and the healthcare system.5 ,9 ,12 However, the inclusion of non-time-loss injuries may be less reliable and further complicate comparisons between studies.10 ,13 This is particularly true when comparing injury incidence from US surveillance systems, which use a variety of injury definitions that often differ from European studies which may have adopted consensus definitions.
Despite this ongoing discussion in US surveillance systems, non-time-loss injuries present an inherent burden of currently unknown magnitude on athletes and medical staff.12 There remains a need to investigate the incidence and distribution of soccer injuries as a whole, including non-time-loss injuries. Sex differences in injury patterns of soccer injuries have been well documented, specifically regarding the higher incidence of anterior cruciate ligament (ACL) and head injuries among female athletes.14–16 However, these discussions have primarily involved specific time-loss injuries, while potential sex differences in non-time-loss injuries have not been fully explored.
This study investigated the incidence and distribution of all reported injuries in NCAA men's and women's soccer athletes. By including non-time-loss injuries in this investigation, new information can be learnt about risk factors for injury among these athletes. These results can also be used to identify specific targets for intervention and to better allocate sports medicine resources in order to limit injury in this extremely popular sport.
Data originate from the NCAA Injury Surveillance Program (NCAA-ISP), which is managed by the Datalys Center for Sports Injury Research and Prevention (Datalys Center), an independent, non-profit research organisation. Men's and women's soccer data were collected from the 2009/2010–2014/2015 academic years. The methodology of the NCAA-ISP has been previously described and is summarised below.17
The NCAA-ISP depends on a convenience sample of NCAA varsity sport teams with athletic trainers (ATs) reporting injury and exposure data. ATs from participating programmes reported information through their electronic health record application throughout the academic year. Common data elements, including injury and exposure information, were deidentified, recoded and exported to an aggregate database. Additionally, the NCAA-ISP captured other sport-related adverse health events such as heat-related conditions, general medical illness and skin infections. Only varsity-level practice and competition events were included in the NCAA-ISP datasets. Junior varsity programmes, as well as any individual weight-lifting and conditioning sessions, were excluded.
For each injury or illness event, the AT completed a detailed report on the injury or condition (eg, body part, diagnosis), circumstances (eg, activity, mechanism) and event type (ie, competition or practice). ATs were able to view and update previously submitted information as needed throughout the season. Additionally, ATs provided the number of student-athletes participating in each practice and competition.
Only deidentified relevant variables and values were exported to the Datalys Center which then passed through an automated verification process that conducted a series of consistency checks.17 The automated verification process would notify the AT and data quality assurance staff of any invalid values or incomplete entries, and staff would assist the AT in the resolution of the issue. Data that passed the verification process were then placed into aggregate datasets for research purposes.
A reportable injury in the NCAA-ISP was defined as an injury that (1) occurred as a result of participation in an organised NCAA-sanctioned practice or competition and (2) required attention from an AT or physician. Multiple injuries occurring from one injury event could be included. Compared with previous NCAA-reported data that only reported time-loss injuries (injuries resulting in participation restriction time of at least 24 h), this 6-year dataset also includes non-time-loss injuries (injuries resulting in participation restriction for <24 h). We classified severe injuries as those resulting in participation restriction time of over 28 days or the student-athlete prematurely ending the season.18
A reportable AE was defined as one student-athlete participating in one NCAA-sanctioned practice or competition which exposed the athlete to the possibility of injury, regardless of the time associated with that participation. Only athletes with actual playing time in a competition, including warm-ups, were included in competition exposures.
Body parts injured were categorised as head/face, neck, shoulder, arm/elbow, hand/wrist, trunk (including chest, abdomen, upper back and lower back), hip/groin, upper leg, knee, lower leg, ankle, foot and other. Diagnoses were categorised as sprain, strain, contusion, concussion, fracture, dislocation, inflammation, laceration, spasm, tendonitis and other. Injury mechanism was defined as the manner in which the student-athlete sustained his/her injury. In the NCAA-ISP, ATs selected from a preset list of options, including player contact, surface contact, equipment contact, contact with out-of-bounds object, non-contact, overuse, illness, infection and other/unknown.
Data were analysed using SAS Enterprise Guide software (V.5.1; SAS Institute Inc, Cary, North Carolina, USA) to assess rates and patterns of collegiate men's and women's soccer injuries. Injury rates were calculated per 1000AE. Statistical analyses included calculation of rate ratios (RRs) and injury proportion ratios (IPRs). The following is an example of an RR comparing practice injury rates in soccer between females and males:
The following is an example of an IPR comparing the proportion of injuries that were severe between females and males:
All 95% CIs not containing 1.0 were considered statistically significant. This study was approved by the National Collegiate Athletic Association Research Review Board.
Overall incidence and rates
Men’s soccer. ATs reported 1554 injuries across 104 team-seasons from 44 programmes over 6 years. A total of 827 (53.2%) injuries occurred during practice and 727 (46.8%) during competition (table 1). Most injuries occurred in the regular season (66.5%, n=1034); 26.6% (n=414) and 6.8% (n=106) occurred in the preseason and postseason, respectively. In total, 47.2% (n=733) were non-time-loss injuries, and 50.7% (n=788) were time-loss injuries, of which 80 (5.1% of all injuries) were severe. These 1554 injuries occurred during 192538AE, for an injury rate of 8.07/1000AE (95% CI 7.67 to 8.47). The time-loss injury rate was 4.09/1000AE (95% CI 3.81 to 4.38). The competition rate was higher than the practice rate for all injuries (17.53 vs 5.47/1000AE; RR=3.20; 95% CI 2.90 to 3.54) and time-loss injuries (9.24 vs 2.68/1000AE; RR=2.99; 95% CI 2.76 to 3.25).
Women’s soccer. ATs reported 2271 injuries across 167 team-seasons from 64 programmes over 6 years. A total of 1161 (51.1%) injuries occurred during practice and 1110 (48.9%) during competition (table 1). Most injuries occurred in the regular season (65.0%, n=1477); 29.4% (n=668) and 5.5% (n=126) occurred in the preseason and postseason, respectively. In total, 47.5% (n=1079) were non-time-loss injuries and 50.8% (n=1153) were time-loss injuries, of which 208 (9.2% of all injuries) were severe. These 2271 injuries occurred during 269112AE, for an injury rate of 8.44/1000AE (95% CI 8.09 to 8.79). The time-loss injury rate was 4.28/1000AE (95% CI 4.04 to 4.53). The competition rate was higher than the practice rate for all injuries (17.04 vs 5.69/1000AE; RR=2.99; 95% CI 2.76 to 3.25) and time-loss injuries (8.86 vs 2.82/1000AE; RR=3.30; 95% CI 2.91 to 3.75).
Sex differences. The injury rates for men and women did not differ in competitions (17.53 vs 17.04/1000AE; RR=1.03; 95% CI 0.94 to 1.13; table 1) or practices (5.47 vs 5.69/1000AE; RR=0.96; 95% CI 0.88 to 1.05). The only statistically significant difference in injury rates was found in preseason practices, where the rate was higher in women than men (10.10 vs 8.57/1000AE; RR=1.18; 95% CI 1.04 to 1.34). Additionally, the proportion of injuries that were severe was higher in women than men (9.2% vs 5.1%; IPR=1.78; 95% CI 1.39 to 2.28).
Body parts injured
The largest proportion of injuries occurred to the lower extremity (hip/groin, upper leg, knee, lower leg, ankle and foot) in competitions (men: 74.1%, n=539; women: 68.0%, n=755) and practices (men: 65.2%, n=539; women: 77.7%, n=902; table 2). Compared with women, men had higher injury rates to the hip/groin in competition (1.86 vs 0.84/1000AE; RR=2.20; 95% CI 1.56 to 3.11), upper leg in competition (3.01 vs 1.98/1000AE; RR=1.52; 95% CI 1.19 to 1.95) and hip/groin in practice (0.80 vs 0.57/1000AE; RR=1.41; 95% CI 1.09 to 1.82). Compared with men, women had higher injury rates to the head/face in competition (3.06 vs 2.19/1000AE; RR=1.39; 95% CI 1.09 to 1.78), knee in competition (3.44 vs 2.15/1000AE; RR=1.60; 95% CI 1.25 to 2.05) and knee in practice (0.89 vs 0.70/1000AE; RR=1.28; 95% CI 1.01 to 1.63). The body parts with the largest proportion of severe injuries in men were the shoulder in competition (20.0%, n=6) and the knee in practice (13.3%, n=14); in women, the largest proportions of severe injuries were in the knee in competition (33.9%, n=76) and the arm/elbow in practice (27.3%, n=3).
The largest proportions of injuries were diagnosed as sprains, strains and contusions in competition and practice for men and women (table 3). Compared with women, men had higher injury rates of strains in competition (3.88 vs 2.58/1000AE; RR=1.51; 95% CI 1.21 to 1.87), lacerations in competition (0.60 vs 0.28/1000AE; RR=2.18; 95% CI 1.19 to 4.00) and contusions in practice (0.68 vs 0.45/1000AE; RR=1.51; 95% CI 1.14 to 2.00). Compared with men, women had a higher concussion rate in competition (1.83 vs 0.94/1000AE; RR=1.94; 95% CI 1.35 to 2.79). The diagnosis with the largest proportion of severe injuries was fractures in men (competition: 44.8%, n=13; practice: 28.6%, n=4) and women (competition: 44.1%, n=15; practice: 33.3%, n=9).
The largest proportions of injuries were due to player contact in competitions (men: 51.2%, n=372; women: 51.7%, n=574) and non-contact in practices (men: 32.0%, n=265; women: 32.6%, n=378; table 4). Compared with women, men had a higher player contact injury rate in practice (1.34 vs 0.83/1000AE; RR=1.60; 95% CI 1.31 to 1.97). Compared with men, women had a higher overuse injury rate in practice (1.31 vs 1.01/1000AE; RR=1.30; 95% CI 1.07 to 1.59). The injury mechanism with the largest proportion of severe injuries was ball contact in men's competition (15.9%, n=7), men's practice (7.2%, n=6), women's practice (8.2%, n=9) and non-contact in women's competition (16.7%, n=34).
The most common injuries included ankle sprains, upper leg strains, hip/groin strains, knee sprains and concussions (table 5). Compared with women, men had a higher hip/groin strain rate than women (0.78 vs 0.46/1000AE; RR=1.69; 95% CI 1.33 to 2.14). Compared with men, women had higher concussion (0.59 vs 0.34/1000AE; RR=1.76; 95% CI 1.32 to 2.35) and knee sprain (0.66 vs 0.42/1000AE; RR=1.59; 95% CI 1.22 to 2.07) rates. In particular, the injury rate for concussions caused by ball contact among women was 2.43 times that of men (95% CI 1.37 to 4.33).
Most epidemiologic research regarding US collegiate soccer injuries has been limited to time-loss injuries.10 ,17 ,19 ,23 This has also been true in studies of European soccer injuries.20 Time-loss injury rates of 20.4/1000 h participation in match and 3.9/1000 h in training have been reported among Dutch amateur soccer players,21 and rates of 7.1/1000 h in match and 1.2/1000 h in training among U-19 Portuguese amateur soccer players.22 However, these studies do not report medical attention or non-time-loss injuries.22 Consequently, these studies do not account for all injuries occurring to soccer athletes, as we found that non-time-loss injuries accounted for nearly half of all injuries requiring medical attention and treatment from team medical staff.
It should be noted that in the consensus definitions proposed by Fuller et al,11 and used in the European studies above, time-loss injuries include those that result in time loss of <1 day, which is a subtle difference from the time loss definition used in US systems which require the injury to result in time loss greater or equal to 1 day.17 This will affect any comparisons between soccer injury studies which have adopted these consensus definitions and those which have not.8 However, universal adoption of such consensus definitions may be difficult to achieve as individual surveillance systems may serve a variety of purposes and function within constraints specific to their subject and data collector populations.5 Further development of injury surveillance should look to merge methods and definitions to improve comparability between studies.
The time-loss injury rates reported in this study are smaller than those previously reported from US studies from the 1988/1989–2003/2004 years in competition (men: 9.24 vs 18.8/1000AE; women: 8.86 vs 16.4/1000AE) and practice (men: 2.68 vs 4.3/1000AE; women: 2.82 vs 5.2/1000AE).3 ,4 The smaller time-loss injury rates in this study may indicate that US collegiate soccer is now safer, or injury prevention programmes have been effective. It may also be an artefact of evolving methods of reporting and surveillance between 2007 and 2015.17 Regardless, the current inclusion of non-time-loss injuries presents a more accurate measure of injury incidence and assists with a more realistic estimation of the burden of total soccer injuries at this time.
In studies using recent NCAA-ISP data, non-time-loss injuries comprised a range of 58.8% (women's cross country) to 77.2% (men's swimming and diving) of all reported injuries.24 ,25 These percentages of non-time-loss injuries are markedly higher than what we found in our investigation. However, the differences between soccer, and swimming and diving and cross country may be due to the nature of these sports, as swimming and cross country depend more on repetitive and cyclic motion. This can predispose athletes to injuries which result in altered participation rather than absence from participation, thus inflating the percentage of injuries which were non-time loss.26
In a study of all athletes, in all sports at one university over 3 years, Yang et al 6 reported that non-time-loss injuries accounted for 50.8% of overuse injuries and 29.8% of acute injuries, equating to roughly 35% of total injuries. Yang et al 6 used a similar non-time-loss definition as the current study; however, non-time-loss incidence was not reported by individual sport. In a study of Danish adolescent soccer, Clausen et al 27 found that 36.6% of self-reported adolescent female soccer injuries were non-time loss as well. A different study of Spanish soccer injury reported that among recreational players under age 30, 68.1% were time loss and 31.9% were medical attention.28 Our higher proportion of non-time-loss injuries (near 50%) may be due to actual differences, but may also be related to differences in injury definition (non-time loss vs medical attention), data collection or improvements in the reporting and representation of non-time-loss injuries.29
The differences in the incidence of non-time-loss injuries between data sources underline the need for the implementation of a common injury definition or definitions within injury surveillance. This is specifically important for non-time-loss injuries which are traditionally harder to standardise and report.5 ,8 Non-time-loss injuries may also be underestimated, specifically if they are not reported to ATs or within injury surveillance.12 ,26 The implementation of a consensus definition will assist with the generalisability and comparability for all injuries across studies, and improve our understanding of injury incidence.
Injury rates in this study did not differ between men's and women's soccer overall or in strata of time-loss and non-time-loss injuries. This is inconsistent with the literature. Previous studies found that the collegiate soccer time-loss injury rate was higher in women than men during practices (5.23 vs 4.34/1000AE), but not competitions (16.44 vs 18.75/1000AE).3 ,4 In high school soccer, compared with boys, girls had a lower time-loss practice injury rate (1.21 vs 1.51/1000AE), but a higher time-loss competition injury rate (5.34 vs 4.26/1000AE).30 These inconsistencies may be due to the period of data collection, as previous studies used data from 1998/1999 to 2002/2003, during which time the number of women's soccer programmes more than tripled, while men's programmes increased by only 30%.3 ,4 Sex differences found in earlier studies may have been due to the rapid growth of women's soccer, where less experienced female players and teams may have been more prone to injury. Injury incidence may have stabilised as these programmes became more established.
We also found that women's soccer athletes had a significantly higher proportion of severe injuries than men's soccer. While this ratio is consistent with the literature, the percentage of severe injuries in this study (9.2% vs 5.1%) was much lower than the proportions found in Belgian youth soccer players (25.3% of female injuries vs 24.1% of male injuries).31 This drastic difference is likely due to differing injury definitions, as the Belgian study defined severe injuries as any fracture, cartilage problem, tendon rupture, concussion, ligament injury or dislocation, compared with our definition of time loss of >28 days.
Our results concur with previous findings that the majority of men's and women's soccer injuries occur to the lower extremity.3 ,4 ,22 ,30 Practice injury rates for men's and women's soccer were highest in the upper leg—highest for the ankle in men, and the knee in women in competition, which is consistent with literature. In collegiate soccer, the risk of hip/groin injury was twice as high in men as women; in high school soccer, the competition knee injury rate was over twice as high in girls as boys.32 ,33 Differences in injury location may be due to anatomical and neuromuscular differences in the lower extremity, and/or differences in training or game play between the sexes.34–36 Future prevention strategies should focus on hip/groin injuries in male players and knee injuries in female players.
There were minimal sex differences in the distribution of the individual diagnoses between men and women, with the exception of strains, contusions and concussions. The higher concussion rate in women is consistent with the literature.15 ,37–39 Of note is that women had a higher rate of concussions from ball contact. Previous studies have attributed this to weaker neck musculature in female soccer athletes, and different levels of contact.40
The use of a voluntary cohort for the NCAA-ISP may result in a potentially unbalanced representation of soccer teams by region or division. Our study did not examine the potential implications of this imbalance, or the effects of variations in AT staffing or injury prevention interventions. Small cell sizes for rare diagnoses, mechanisms and body parts injured resulted in some imprecise measurements of rates. AEs do not account for the amount of time in which each individual athlete participates in each event.
The higher proportion of time-loss compared with non-time-loss injuries may be an artefact of previous surveillance formats, which only allowed the reporting of time-loss injuries.17 ATs who had been participating in the NCAA-ISP prior to the inclusion of non-time-loss injuries may not have reported all non-time-loss injuries due to habit. It is also possible that some non-time-loss injuries progressed into time-loss injuries and were only reported to the NCAA-ISP as time-loss injuries.26 Therefore, the non-time-loss incidence may be artificially lowered due to reporting bias. Time-loss injuries may also be considered ‘more important’, more intensive for ATs to treat, and therefore have a greater likelihood of being reported.26 Methodological differences in data collection among US studies and between US and European studies limit direct comparisons of soccer injury incidence between separate populations. The overall adoption and implementation of consensus definitions would greatly assist with such comparisons.
We found that overall, non-time-loss injuries account for nearly half of the total soccer injuries, as well as sex differences in the distribution of specific injury characteristics. These differences were most pronounced in time-loss competition injuries, which occurred at higher rates in men than in women, and sex differences in head injuries and concussion, specifically during competition. Further investigation into these differences is warranted. The overall high percentage of non-time-loss injuries indicates that a large proportion of injuries to collegiate soccer players may not be as well understood. These measures of non-time-loss injuries may also be an underestimation due to under-reporting and the lack of a common definition implemented within injury surveillance. Further study into the incidence, treatment and outcome of non-time-loss injuries, as well as the overall implementation of a common injury definition, may identify a more accurate burden of these injuries on performance and subsequent injury.
What are the findings?
The study updated previous findings related to US collegiate soccer players by examining time-loss and non-time-loss injuries.
Non-time-loss injuries account for nearly half of the injuries in collegiate soccer players.
Female soccer players were more likely to sustain concussions in competitions than male soccer players.
Prevention interventions should focus on hip/groin injuries in male soccer players and knee injuries in female soccer players.
The NCAA Injury Surveillance Program data were provided by the Datalys Center for Sports Injury Research and Prevention. The Injury Surveillance Program was funded by the NCAA. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the NCAA. The authors thank the many ATs who have volunteered their time and efforts to submit data to the NCAA Injury Surveillance Program. Their efforts are greatly appreciated and have had a tremendously positive effect on the safety of collegiate athletes.
Contributors KGR led the analysis of the study and drafting of the manuscript. ZYK led the creation and analysis of the study. SLD and AD assisted with the analysis of the study, interpretation of findings and drafting of the manuscript. The remaining authors, EBW, AG and TPD, assisted with the creation of the study and drafting of the manuscript. All authors approve the final manuscript as presented.
Funding The NCAA Injury Surveillance Program data were provided by the Datalys Center for Sports Injury Research and Prevention. The Injury Surveillance Program was funded by the NCAA.
Ethics approval National Collegiate Athletic Association Research Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.