Share this article on:

Injuries in Children with Extra Physical Education in Primary Schools

Trifonov Rexen, Christina1; Andersen, Lars Bo1,2; ErsbØll, Annette KjÆr3; Jespersen, Eva1; Franz, Claudia1; Wedderkopp, Niels1,4,5

Medicine & Science in Sports & Exercise: April 2014 - Volume 46 - Issue 4 - p 745–752
doi: 10.1249/MSS.0000000000000152

Purpose (1) Examine the influence of extra physical education (EPE) on the number of musculoskeletal injuries in public schools accounting for organized sports participation (OSP) outside school. (2) Examine the major injury subgroup: growth-related overuse (GRO) through the overuse-related injury group.

Methods A longitudinal controlled school-based study among Danish public schools. At baseline, 1216 children participated age 6.2–12.4 yr. Six schools (701 children) with EPE and four control schools (515 children) were followed up with weekly automated mobile phone text messages for information on musculoskeletal problems and OSP. Health care personnel diagnosed the children according to the World Health Organization’s International Classification of Diseases, Tenth Revision. Data were analyzed using a two-part zero-inflated negative binomial (ZINB) regression model.

Results School type had no influence on the odds of sustaining an injury but increased the probability of sustaining a higher injury count for children with injuries, with total injuries by a factor of 1.29 (95% confidence interval [CI] = 1.07–1.56), overuse by a factor of 1.29 (95% CI = 1.06–1.55), and GRO by a factor of 1.38 (95% CI = 1.02–1.80). Weekly mean OSP decreased the odds of belonging to the group of children with no injuries, by a factor of 0.29 (95% CI = 0.14–0.58), 0.26 (95% CI = 0.14–0.48), and 0.17 (95% CI = 0.06–0.52) for total, overuse, and GRO, respectively. OSP also increased the probability of sustaining a higher injury count for children with injuries by a factor of 1.11 (95% CI = 1.02–1.22), 1.10 (95% CI = 1.00–1.22), and 1.14 (95% CI = 1.00–1.30), respectively.

Conclusions Children enrolled in EPE schools with high OSP have the highest odds of injury and a high probability of sustaining a higher injury count compared to their peers at schools with normal PE. Special attention should be assigned to these children during compulsory PE.

Supplemental digital content is available in the text.

1Centre of Research in Childhood Health, Institute of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, DENMARK; 2Department of Sports Medicine, The Norwegian School of Sport Sciences, Oslo, NORWAY; 3National Institute of Public Health, University of Southern Denmark, Copenhagen, DENMARK; 4Sport Medicine Clinic, Orthopedic Department, Hospital of Lillebaelt, Odense, DENMARK; and 5Institute of Regional Health Services Research, University of Southern Denmark, Odense, DENMARK

Address for correspondence: Christina Christiansen, Chiropractor, Centre of Research in Childhood Health, Institute of Sports Science and Clinical Biomechanics, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark; E-mail:

Submitted for publication June 2013.

Accepted for publication August 2013.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (

Health-related benefits of regular engagement in physical activity (PA) in the population are widely known and include decreased risk of cardiovascular disease, obesity, diabetes mellitus type 2, and osteoporosis. Physical education (PE) in school is an important arena for PA because it is compulsory in most countries and therefore includes the least active children (16). In Denmark, the number of schools with extra amount of PE, referred to as “EPE schools,” is increasing. In May 2011, 38 EPE schools (2.8%) existed and spread across 18 municipalities in Denmark. The total number of students in those schools was 10,315, an increase of 24% compared to May 2010 (33). Although this setting is specific to Denmark, the applicability of an increased amount of PE in a school setting would be possible in most countries.

A major drawback of enlarged engagement in sports and PA is the increased risk of sports-related injuries, both during PE classes in school and during participation in organized sports and PA during leisure time (36). It has been speculated that an increasing number of children undertake intensive physical training at younger ages or participate in multiple sports, increasing the overall exposure. In some cases, this leaves too little time for tissue remodeling and growth, leading to an increased risk of acute and overuse injuries (1,13). A polarization in sports activity between children in Denmark was seen in the period 1997 to 2007, with an overall decline in children participating in sports but an increase in time spent in sports among the sports active children (30).

Overuse injuries have been reported to account for up to 50% of all injuries in children and adolescents (10,38). Growth-related overuse (GRO) injuries are a major subgroup of injuries in growing individuals (19,26). Osgood–Schlatter’s disease, Sinding-Larsen–Johansson’s disease, and Sever’s disease are examples of typical GRO injuries to the apophyseal center of ossification, at the point of attachment of a major musculotendinous unit to bone, also called traction apophysitis (25,32). Traction apophysitis usually occurs during periods of rapid growth, when the growth and ossification of cartilage occurs at locations involved in considerable and repetitive tension forces from pulling muscles (26,28). Overuse injuries in children and adolescents may be the result of training error, excessive training with increased intensity and inadequate rest, the growth process combined with repetitive training, improper technique, muscular imbalance, participation in multiple sports, or early specialization (5,13,26). Introducing EPE schools may prevent injuries through enhanced levels of fitness and motor performance. However, the increased overall exposure considering extra PE lessons in school and sports participation outside school could possibly increase the risk of injuries.

Sports-related injuries have negative short- and long-term effects. On the short term, injuries lead to pain, disability, and decreased socialization and participation in sports. Injuries related to joints, e.g., cartilage, ligaments, and joint capsules, increase the risk of developing secondary osteoarthritis, permanent loss of function, absence from work later in life, and significant health care cost (20,34).

Increasing the amount of PE lessons in public schools has several important potential health-related benefits. However, increasing the amount of PE lessons in public schools may have ethical considerations. The overall outcome of interest is improved health; therefore, the drawbacks must be minor and exceeded by improved health for the continuation and expansion of public schools with extra PE lessons. In this setting, injuries are the obvious concern. If increased amount of PE lessons shows increased risk of injuries, then the severity and extent have to be evaluated to ensure the ethical concerns of the concept.

Back to Top | Article Outline


This study aimed to investigate the influence of EPE in public schools on the number of injuries, adjusted for participation in organized sports to account for the total organized exposure, for each child and to further investigate the major injury subgroup: GRO injuries through the injury group of overuse-related injuries.

Back to Top | Article Outline


Study population

Ten public schools in the municipality of Svendborg, Denmark, were observed. Six schools had EPE (EPE) with 270 min·wk−1 and four schools had the normal amount of 90 min·wk−1. Schools with EPE were planned by the municipality, in cooperation with the schools, and were matched with four schools with normal physical education (NPE) based on size and geography. This study was a substudy of “The Childhood Health, Activity, and Motor Performance School Study Denmark” (The CHAMPS-study DK) (39). The children were followed up for 2 yr. In August 2009, 1216 children participated (NPE = 515, EPE = 701). The project was continuously open for new participants.

Back to Top | Article Outline

PE lessons in EPE schools

To ensure the optimal quality of the PE lessons, PE teachers and pedagogues participated in a 40-h course in “age-related concepts of training” by Team Denmark, which was held to ensure targeted and responsible training for the growing individual (31). Furthermore, a handbook was provided with inspiration for the content of PE lessons with practical suggestions and exercises (3).

Back to Top | Article Outline

Short messaging service

Injuries and organized sports participation (OSP) were derived from a weekly “Short Messaging Service (SMS) Questionnaire” (SMS-track, New Agenda Solutions version 2.1) (27). The method functions as a “follow-up” procedure and is used to investigate the fluctuations in complaints from the musculoskeletal system and sports participation over time. Parent reports were used as proxy for the children’s response and considered appropriate in this cohort, as self-report questionnaire in young children has shown to be unreliable (4). Previous studies have shown SMS-track to be more reliable and cheaper compared to repeated paper-based surveys with recall over longer periods (18). A Swedish study among patients with lower back pain, age 16–69 yr, has found the SMS-track system to be user-friendly with high response rates (82.5%) and unaffected by season (2).

The questions used in the present study were automatically sent to the parent’s mobile phone at the end of each week asking two questions on different SMSs:

Question 1: “Has (the child’s name) during the last week had any pain?” — with answer possibilities of 1 = neck, back, or low back; 2 = shoulder, arm, or hand; 3 = hip, leg, or foot; or 4 = no, my child has not had any pain.

Question 2: “How many times did (the child’s name) engage in sports during the last week?” — with answer possibilities of 0 = none, 1 = once, 2 = twice, …, or 8 = more than seven times.

The returned answers were automatically recorded in a database distributed by SMS-track, New Agenda Solutions (27). To improve compliance rate, a reminder was automatically sent if participants had not responded within 72 and 120 h after receiving the first message. Furthermore, the respondents were contacted by phone if the answer did not meet the instructions.

The answer rate for SMS-track question 1 was 92.19% (5.75% missing answers, 1.63% technical errors, 0.36% invalid answers, and 0.07% empty answers). For SMS-track question 2, the answer rate was 92.10% (5.52% missing due to unanswered question 1, 1.58% technical error, 0.53% missing answers, 0.22% invalid answers, and 0.05% empty answers).

Back to Top | Article Outline

Data on injuries

Data on injuries as the number of International Classification of Diseases, Tenth Revision (ICD-10; World Health Organization) diagnoses were collected based on SMS-track. Parents of children reporting any pain in the past week were contacted by phone by researchers with health care education (physiotherapists or chiropractors). The child was examined if an injury was possible. Occasionally, the child had already been, or was to be, examined by other health care units, and the information was then collected (e.g., emergency departments, specialists, or hospitals). The injury definition in this study was any ICD-10 diagnosis related to the musculoskeletal system.

The term “GRO injuries” is used for the group of nontraumatic injuries observed in growing individuals related to growth zones in extremities. The World Health Organization’s International Classification defines this injury group “chondropathies” comprising the ICD-10 groups M91–M94 (40). From those, injuries around the knee and heel show similar etiology with traction apophysitis of the calcaneus, patella, and tuberositas tibiae.

Back to Top | Article Outline

Data on organized sports participation

OSP in leisure time was collected through SMS-track. Analyses were corrected for OSP in leisure time because participation was expected to play a significant role in the exposure of possible injuries.

Back to Top | Article Outline


The study was carried out in accordance with the Declaration of Helsinki and was approved by the local scientific ethics committee (ID S20080047) and registered in the Danish Data Protection Agency ( 2008-41-2240). Parents of the children gave written informed consent, and children gave verbal consent.

Back to Top | Article Outline

Outcome measures

In this study, the total number of injuries experienced for each child was of interest. Diagnoses made by the research team with health care background and from other health care units were summarized for each child and further divided into subgroups with counts of overuse and GRO injuries. GRO consisted of calcaneal apophysitis, Osgood–Schlatter, and Sinding-Larsen–Johansson disease (ICD-10 M92.8, M92.5, and M92.4, respectively).

Back to Top | Article Outline

Statistical analysis

A total of 1310 children participated in the study, of whom 1216 were included at baseline (i.e., August 2009) and 94 were included by drop-in. In total, 77 children dropped out during the study (EPE = 50, NPE = 27). Sixty-five children dropped out because of a change in school. SMS-track information on OSP was available in 1231 children, and only 3 children had invalid answers. To account for children with a low grade of information, either because of low answer rates or partial participation due to drop-in or dropout, a participation rate of 60%, of the possible 87 weekly answers on OSP, was used as the minimum for inclusion in the analyses. This led to exclusion of 166 children (NPE = 66, EPE = 100, corresponding to 13% and 14% of all children at NPE and EPE schools, respectively), 108 of which only had partial participation.

Analyses were performed for the following outcomes: total injury count, overuse-related injury count, and GRO injury count with school type as primary explanatory variable and OSP, sex, and grade level at baseline as covariates. OSP was used as the mean number of sports participation per week for each child. The sample size was considered adequate (see text document, Supplemental Digital Content 1,, which demonstrates preanalysis power calculation). The number of injuries was heavily skewed to the right and contained a high proportion (40.3%, 50.5%, and 69%) of zero counts (i.e., excess zeros). Initially, a graphical visualization was performed, investigating how well the outcomes fitted both a Poisson and a negative binomial distribution. Overall, the negative binomial distribution fitted data better than the Poison distribution due to overdispersion (i.e., the mean and variance were not equal but differed by a factor of 1.3 to 1.9). The data structure was hierarchical with three levels, that is, children nested within classes nested within schools. Initially, a three-level Poisson regression model for each of the three outcome measures (number of injuries) was fitted with school, classes within school, and children within classes as random effects. The baseline grade level was included as a fixed effect at class level, and school type (EPE and NPE) was included as a fixed effect at school level. The models had a poor fit due to the high proportion of zero counts and unequal mean and variance in data. Data with excess zeros have been recognized as a problem because of the influential effect of the zeros versus the nonzeros. Zero-inflated count models provide an alternative to analyzing count data with dispersion and excess zeros (22). Count outcomes, such as number of injuries, are often characterized by a large proportion of zeros and inequality between mean and variance. Zero-inflated count models [as the zero-inflated negative binomial model (ZINB)] are simultaneously modeling factors associated with risk of injuries (no vs one or more injury) and factors associated with the number of injuries among those who have one or more injury. Alternatives to this modeling approach include a logistic regression of injuries categories as a binary variable (injury: yes or no) with loss of information regarding the number of injuries for students with more than one injury. The zero-inflated models can examine the effect of the explanatory variables and covariates in a model, simultaneously modeling the risk of getting an injury and the number of injuries. A ZINB model was fit to the data. The multilevel data structure was accounted for by using a robust variance estimator at the class level. Data were analyzed using STATA (version 11.1) with a 5% significance level. Backward elimination of nonsignificant covariables was performed. However, these models were compared with alternative models because backward elimination is problematic in zero-inflated models favoring the inflate part over the negative binomial part. Model choice was based on plots visualizing differences between observed and predicted values, choosing the model with the best model fit supported with the Voung test (23,37). Two interactions were found in the analyses of GRO injuries: first, between-sport mean and grade levels in the inflate part of the model giving a more complicated model without improved model fit or interpretation; and second, between-sport mean and sex in the negative binomial part of the model leading to a decrease in model fit without notably added information.

Back to Top | Article Outline


Table 1 and Figure 1 summarize the analyzed study population. In total, 1062 children had 1416 injuries, including 1005 overuse-related injuries, of which 454 were GRO injuries. The number of students with injury count of zero to five or more was as follows: 428 (40.3%), 262 (24.7%), 158 (14.9%), 110 (10.4%), 55 (5.2%), and 49 (4.7%). Participation in organized sports outside school (Fig. 2) was surprisingly low at EPE schools, underlining the importance of adjusting the statistical models for this effect when investigating school type differences in injuries to account for the overall exposure from both inside and outside the school. Coefficients for the final ZINB models are shown in Table 2, separated into the logistic and the negative binomial part of the model. The logistical part of the model provides information on the odds of belonging to the group of children without injuries compared to the group of injured children in relation to explanatory variables. The negative binomial part of the model provides information of the injury count among the children with injuries in relation to explanatory variables. The children with no injuries are therefore not considered in the negative part of the model.









Back to Top | Article Outline

Total injuries and overuse-related injuries

School type had no effect on the odds of getting injured or not (logistic part of the models). For each additional OSP per week, the odds of belonging to the group of children with no injuries decreased by a factor 0.29 and 0.26, respectively, holding all other variables constant (total: 0.29 [95% confidence interval [CI] = 0.14–0.58]; overuse: 0.26 [95% CI = 0.14–0.48]). In other words, OSP increased the risk of injury unaffected by school types.

Examining the group of injured children (the negative binomial part of the models), EPE school type increased the number of total and overuse injuries equally with a factor: 1.29 (total: 95% CI = 1.07–1.56; overuse: 95% CI = 1.06–1.55). OSP increased the expected injury count by almost the same level for the total injury count and for overuse injuries (total: 1.11 [95% CI = 1.02–1.22]; overuse: 1.10 [95% CI = 1.00–1.22]). Girls had a higher injury count than boys (total: 1.22 [95% CI = 1.05–1.42]; overuse: 1.23 [95% CI = 1.05–1.42]).

Back to Top | Article Outline

GRO injuries as subgroup

School type had no effect on the odds of getting a GRO injury or not (logistic part of the model). Again, for each additional OSP per week, the odds of belonging to the group of children with no GRO injury decreased by a factor of 0.17 (95% CI = 0.06–0.52), holding all other variables constant. In other words, OSP also increased the risk of GRO injury unaffected by school types. Examining the group of children with GRO injuries (the negative binomial part of the model), EPE school type increased the number of GRO injuries by a factor of 1.38 compared to their peers at NPE schools (95% CI = 1.05–1.80). OSP was borderline significant (P = 0.051); however, OSP increased the expected injury count almost equally compared to the count models for the total injuries and overuse injuries (GRO injuries: 1.14 [95% CI = 1.00–1.30]). No effect was found for sex.

The difference in accumulated probability of experiencing one or more injuries as weekly OSP varies is shown in Figure 3. A higher mean OSP resulted in a higher probability of injuries. The school type difference is most pronounced for GRO injuries as weekly OSP increases.



Six diagnoses in the category of “ICD-10 Chondropathies” were not analyzed together with the GRO injuries, which only counted apophysitis at the heel and knee. The six injuries consisted of three children with M93.9 “Osteochondropathy, unspecified,” two with osteochondrosis related to the upper extremities (M92.0 and M92.2), and one with juvenile osteochondrosis of the pelvis (M91.0, iliac crest). All six additional osteochondroses were self-limiting.

Back to Top | Article Outline


During the 2-yr period, 59.7% of the children experienced at least one injury. The proportion of children with overuse-related injuries was consistent with earlier estimations of overuse-related injuries, estimated to account for approximately 50% of all pediatric injuries (35). GRO injuries to the heel and knee were identified to be the largest subgroup of all injuries with a similar etiology (32% of all injuries). Although not serious compared to fractures, sprains, and more severe GRO injuries, any injury is considered important in a health perspective and in the prevention of lifestyle-related diseases through education in an active lifestyle (7,8,11,14,15,24). Schools with EPE showed no injury preventive effect, neither did it increase the odds of belonging to the group of children with injuries, but EPE in school increased the risk of a higher injury count among the relatively large group of children experiencing one or more injuries. NPE schools had a remarkable higher OSP compared to schools with EPE. This fact can give rise to other questions: Is OSP outside school injury preventive and/or does EPE in schools increase the risk of injuries? This study can only answer the overall effect of introducing schools with EPE compared to schools with normal PE.

Sports involving a high rate of physical contact, jumping, sprinting, and pivoting have been shown to cause injuries in children and adolescents (6,12). Other studies have shown a preventive effect in injuries among school-age children. Common for those are a more specific approach toward injury prevention. The majority of injury preventive studies focus on specific sports types. LaBella et al. (21) found that a 20-min neuromuscular warm-up program reduced noncontact lower extremity injuries in female high school soccer and basketball athletes. A more seldom and general approach was made by Collard et al. (9), investigating an 8-month school-based injury prevention program, on all types of PA-related injuries, among children age 10–12 yr, with no overall significant findings. However, children with low activity levels showed an intervention effect. In general, a specific comprehensive warm-up with neuromuscular control, balance, flexibility, strength, technique, and agility drills fitted to sports activity have shown preventive effect in various combinations (17,29,35).

Alternatives to ordinary exercises may be a way to prevent GRO injuries caused by mandatory PE lessons. Differential activities will allow children with present GRO pain on high intensity or prolonged activities to participate in the PE lessons.

With the believed etiology of GRO and general injury prevention in mind, alternative activities with emphasis on technique, coordination, and precision, without repetitive tensile forces through the major muscle groups over the knee and heel, would be a reasonable suggestion. Further research is needed to confirm this suggestion.

Our results identified children with high OSP outside school simultaneously enrolled in schools with EPE to have the highest probability of high injury count. In this context, it is important to be aware of overall exposure. Children are exposed through 1) PE lessons at schools, 2) OSP, and 3) nonorganized PA including activities in school breaks and leisure time. This means that active children who engage in multiple sports or are training at high levels outside school are exposed to an even higher training load than usual and, maybe more important, have less time between activities to recover. Special attention should be placed on these children during the compulsory PE.

An advantage of the present observational study is that it is easily implementable outside the research environment, increasing the transmissibility to everyday life. A possible bias in this study is over-reporting of pain during the weekly SMS questionnaire leading to overdiagnosis of injuries. However, the chiropractors and physiotherapists were aware of this issue, and each ICD-10 diagnosis was made based on the presence of relevant clinical findings only. We assume equal report behavior on injuries between school types based on the very high answer rate of the weekly SMS-track (10.01% missing on NPE schools and 6.17% missing on EPE schools) and an overall mean of 80.7 answer weeks of 87 possible.

Back to Top | Article Outline

Strengths and limitations

A major strength in this study is the weekly repeated measurements of injuries and OSP, which minimize recall bias. Other strengths are the professional evaluation of injuries compared with self-reported injuries increasing the accuracy of injury type, and the high participation and response rates. Limitations are missing information on participation rate in PE lessons and nonorganized leisure time activity. Measurements of PA by accelerometry may increase the information about overall exposure.

Back to Top | Article Outline


Sixty percent of all children experienced one or more musculoskeletal injuries. GRO injuries to the knee and heel accounted for 32% of all injuries and were identified as the largest subgroup of injuries with a similar etiology. Children enrolled in schools with EPE and a high OSP have the highest odds of injury and a high probability of sustaining a higher injury count compared to their peers at schools with the normal amount of PE. Special emphasis should be placed on these children during the compulsory PE in schools.

The authors gratefully acknowledge the work of students and PhD students who have participated in the data collection for the CHAMPS-study DK. The authors thank children, parents, and teachers in the schools involved in the project, and we are grateful for the cooperation with “Elitesekretariatet” in the municipality of Svendborg.

This study was supported by grants from The IMK Foundation, The Nordea Foundation, The TRYG Foundation—all private, non-profit organizations, which supports research in health prevention and treatment, and TEAM Denmark, the elite sports organization in Denmark, that provided the grant for the SMS-track system.

All authors state that they have no competing interest.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Adirim TA, Cheng TL. Overview of injuries in the young athlete. Sports Med. 2003; 33 (1): 75–81.
2. Axen I, Bodin L, Bergstrom G, et al. The use of weekly text messaging over 6 months was a feasible method for monitoring the clinical course of low back pain in patients seeking chiropractic care. J Clin Epidemiol. 2012; 65 (4): 454–61.
3. Bach LG, Eiberg S. Aldersrelateret træning, Håndbog for 0. til 6. Klasse [“Age-Related Training, Handbook Preschool to 6th Grade” ]. 1st ed. Brøndby (Denmark): Formegon ApS, DK; 2010. pp. 1–114.
4. Baranowski T, Smith M, Baranowski J, et al. Low validity of a seven-item fruit and vegetable food frequency questionnaire among third-grade students. J Am Diet Assoc. 1997; 97 (1): 66–8.
5. Brenner JS. Overuse injuries, overtraining, and burnout in child and adolescent athletes. Pediatrics. 2007; 119 (6): 1242–5.
6. Caine D, Caine C, Maffulli N. Incidence and distribution of pediatric sport-related injuries. Clin J Sport Med. 2006; 16 (6): 500–13.
7. Caine D, DiFiori J, Maffulli N. Physeal injuries in children’s and youth sports: reasons for concern? Br J Sports Med. 2006; 40 (9): 749–60.
8. Caine DJ, Golightly YM. Osteoarthritis as an outcome of paediatric sport: an epidemiological perspective. Br J Sports Med. 2011; 45 (4): 298–303.
9. Collard DC, Chinapaw MJ, Verhagen EA, Bakker I, van MW. Effectiveness of a school-based physical activity–related injury prevention program on risk behavior and neuromotor fitness a cluster randomized controlled trial. Int J Behav Nutr Phys Act. 2010; 7: 9.
10. Dalton SE. Overuse injuries in adolescent athletes. Sports Med. 1992; 13 (1): 58–70.
11. Dixon M, Fricker P. Injuries to elite gymnasts over 10 yr. Med Sci Sports Exerc. 1993; 25 (12): 1322–9.
12. Emery CA, Meeuwisse WH, McAllister JR. Survey of sport participation and sport injury in Calgary and area high schools. Clin J Sport Med. 2006; 16 (1): 20–6.
13. FIMS/WHO. Excessive physical training in children and adolescents. A position statement from the International Federation of Sports Medicine (FIMS). Schweiz Z. Sportmed. 1991; 39 (1): 32–4.
14. Flynn JM, Lou JE, Ganley TJ. Prevention of sports injuries in children. Curr Opin Pediatr. 2002; 14 (6): 719–22.
15. Garrick JG, Requa RK. Sports and fitness activities: the negative consequences. J Am Acad Orthop Surg. 2003; 11 (6): 439–43.
16. Heath GW, Parra DC, Sarmiento OL, et al. Evidence-based intervention in physical activity: lessons from around the world. Lancet. 2012; 380 (9838): 272–81.
17. Herman K, Barton C, Malliaras P, Morrissey D. The effectiveness of neuromuscular warm-up strategies, that require no additional equipment, for preventing lower limb injuries during sports participation: a systematic review. BMC Med. [Internet] 2012 [cited 2013 Mar];10. Available from: doi: 10.1186/1741-7015-10-75.
18. Johansen B, Wedderkopp N. Comparison between data obtained through real-time data capture by SMS and a retrospective telephone interview. Chiropr Osteopat. [Internet] 2010 [cited 2013 Mar];18. Available from: doi:10.1186/1746-1340-18-10.
19. Kannus P, Nittymaki S, Jarvinen M. Athletic overuse injuries in children. A 30-month prospective follow-up study at an outpatient sports clinic. Clin Pediatr (Phila). 1988; 27 (7): 333–7.
20. Kujala UM, Kettunen J, Paananen H, et al. Knee osteoarthritis in former runners, soccer players, weight lifters, and shooters. Arthritis Rheum. 1995; 38 (4): 539–46.
21. LaBella CR, Huxford MR, Grissom J, Kim KY, Peng J, Christoffel KK. Effect of neuromuscular warm-up on injuries in female soccer and basketball athletes in urban public high schools: cluster randomized controlled trial. Arch Pediatr Adolesc Med. 2011; 165 (11): 1033–40.
22. Lachenbruch PA. Analysis of data with excess zeros. Stat Methods Med Res. 2002; 11 (4): 297–302.
23. Long JS, Freese J. Regression Models for Categorical Dependent Variables Using Stata. 2nd ed. Collage Station (TX): Stata Press; 2006. pp. 349–414.
24. Maffulli N, Chan D, Aldridge MJ. Derangement of the articular surfaces of the elbow in young gymnasts. J Pediatr Orthop. 1992; 12 (3): 344–50.
25. Medlar RC, Lyne ED. Sinding-Larsen-Johansson disease. Its etiology and natural history. J Bone Joint Surg Am. 1978; 60 (8): 1113–6.
26. Micheli LJ, Fehlandt AF Jr. Overuse injuries to tendons and apophyses in children and adolescents. Clin Sports Med. 1992; 11 (4): 713–26.
27. New Agenda Solutions Web site: SMS-Track Questionnaire 2.1 [Internet]. Denmark [cited 2013 Feb 1]. Available from:
28. Ogden JA, Southwick WO. Osgood-Schlatter’s disease and tibial tuberosity development. Clin Orthop Relat Res. 1976; (116): 180–9.
29. Olsen OE, Myklebust G, Engebretsen L, Holme I, Bahr R. Exercises to prevent lower limb injuries in youth sports: cluster randomised controlled trial. BMJ. [Internet] 2005 [cited 2013 Mar];330(7489). Available from: doi:10.1136/bmj.38330.632801.8F.
30. Pilgaard M. Sport og motion i danskernes hverdag [“Sports and Exercise in Danish Everyday Life”]. Copenhagen (Denmark): Idrættens Analyseinstitut; 2009. pp. 23–42.
31. Pryce R, Willeberg S, Falkentoft C, Meyhoff T. Aldersrelateret træning - Målrettet og forsvarlig træning af børn og unge [“Age-Related Concepts of Training - Targeted and Safe Training of Children and Adolescents” ]. Bøndby (Denmark): Paritas Grafik A/S Team Danmark; 2005. pp. 30–70, 150–86.
32. Saperstein AL, Nicholas SJ. Pediatric and adolescent sports medicine. Pediatr Clin North Am. 1996; 43 (5): 1013–33.
33. Team Denmark Web site [Internet]. Brøndby (Denmark): Team Danmark; [cited 2012 May 7]. Available from:
34. Tveit M, Rosengren BE, Nilsson JK, Karlsson MK. Former male elite athletes have a higher prevalence of osteoarthritis and arthroplasty in the hip and knee than expected. Am J Sports Med. 2012; 40 (3): 527–33.
35. Valovich McLeod TC, Decoster LC, Loud KJ, et al. National Athletic Trainers’ Association position statement: prevention of pediatric overuse injuries. J Athl Train. 2011; 46 (2): 206–20.
36. Verhagen E, Collard D, Paw MC, van Mechelen W. A prospective cohort study on physical activity and sports-related injuries in 10–12-year-old children. Br J Sports Med. 2009; 43 (13): 1031–5.
37. Vuong QH. Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica. 1989; 57 (2): 307–33.
38. Watkins J, Peabody P. Sports injuries in children and adolescents treated at a sports injury clinic. J Sports Med Phys Fitness. 1996; 36 (1): 43–8.
39. Wedderkopp N, Jespersen E, Franz C, et al. Study protocol. The Childhood Health, Activity, and Motor Performance School Study Denmark (The CHAMPS-study DK). BMC Pediatr. [Internet]. 2012 [cited 2013 Mar];12(1):128. Available from: doi:10.1186/1471-2431-12-128.
40. World Health Organization: International Classification of Diseases [Internet]. [cited 2013 Feb 1]. Available from:


Supplemental Digital Content

Back to Top | Article Outline
© 2014 American College of Sports Medicine