UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Community-driven injury prevention in youth female soccer Frew, Kira 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2013_fall_frew_kira.pdf [ 22.63MB ]
Metadata
JSON: 24-1.0074175.json
JSON-LD: 24-1.0074175-ld.json
RDF/XML (Pretty): 24-1.0074175-rdf.xml
RDF/JSON: 24-1.0074175-rdf.json
Turtle: 24-1.0074175-turtle.txt
N-Triples: 24-1.0074175-rdf-ntriples.txt
Original Record: 24-1.0074175-source.json
Full Text
24-1.0074175-fulltext.txt
Citation
24-1.0074175.ris

Full Text

COMMUNITY-DRIVEN INJURY PREVENTION IN YOUTH FEMALE SOCCER  by KIRA FREW  BHK, University of British Columbia, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2013  ? Kira Frew, 2013 ii Abstract  Introduction: Youth female soccer players are at high risk of lower-extremity (LE) injury. Randomized controlled trials (RCTs) have previously demonstrated the efficacy of team-based neuromuscular training in decreasing injury rates in youth female soccer players. In an RCT, the neuromuscular training program used in this study was efficacious in reducing the risk of all injuries by 38% and acute-onset injuries by 43% in youth soccer players. The aim of this thesis was to determine the effectiveness of such an injury prevention program when community initiated, taught and delivered.  Research design: Historical cohort study  Participants: In 2008, 23 teams participated in the collection of quality assurance data (n=351). In 2010, 15 teams completed the study (n=187). Players in both cohorts were ages 9 to 17. Intervention: The program included a team-based neuromuscular training warm-up (including dynamic stretching, strength, agility, plyometric and balance components) and an individual home-based wobble board training program.  Main outcome measures: Soccer injury resulting in time loss of one week or more. A soccer injury was defined as any injury occurring during soccer activity resulting in medical attention and/or the removal of the player from the current session and/or subsequent time loss of at least one soccer session as a direct result of that injury.  Results: In study 1, the Risk Ratios (RR) comparing the intervention season to the control season were: all injury (RR, 0.73; 95% CI, 0.37-1.45), acute-onset injury (RR, 0.69; 95% CI, 0.33-1.44) and LE injury (RR, 0.74; 95% CI, 0.34-1.64). In study 2, there was evidence that player position and right knee flexion-to-extension ratio were significant risk factors for injury in youth female soccer players participating in an injury prevention program. Conclusions: RR point estimates suggest that a community-driven team-based neuromuscular training program may be protective of all injury, acute-onset injury and LE injury in youth female soccer players. The magnitude of this effect is similar to that previously determined in RCT studies examining a similar neuromuscular training program. Future research should focus on the implementation context of delivery of such a program to evaluate adherence and maintenance in a youth soccer population.  iii Preface The research in this thesis received a full prior ethical review and approval by the University of Calgary Conjoint Health Research Ethics Board [Ethics ID: E-22608].  Research conducted at the University of British Columbia (UBC) or by persons affiliated with UBC must receive approval from an UBC-affiliated Research Ethics Board (REB), but there was an oversight and this was not done. The UBC REB does not give retroactive approvals for research. However, in this case, because the research had received prior approval from the University of Calgary, the research team was allowed to submit a full application for ethical review to UBC REB after the fact. The UBC REB reviewed this and was satisfied that the research would have met UBC REB standards, had it been reviewed prior to its conduct. Due to the specific and unusual circumstances of this case, the Vice President of Research at UBC has made a one-time exceptional decision to endorse the University of Calgary?s approval of the research, allowing the student to complete the program requirements and graduate.  iv Table of contents Abstract.................................................................................................................................... ii!Preface..................................................................................................................................... iii!Table of contents .................................................................................................................... iv!List of tables............................................................................................................................. x!List of figures......................................................................................................................... xii!List of abbreviations ............................................................................................................ xiii!Acknowledgements ............................................................................................................... xv!Chapter  1: Introduction ........................................................................................................ 1!Chapter  2: Literature review, objectives and hypotheses.................................................. 4!2.1! Participation in youth soccer........................................................................................ 4!2.2! The epidemiology of soccer-related injuries................................................................ 4!2.2.1! Injury definition .................................................................................................... 5!2.2.2! Injury incidence .................................................................................................... 6!2.2.3! Injury location and type ...................................................................................... 14!2.2.4! Injury severity ..................................................................................................... 18!2.2.5! Injury mechanism................................................................................................ 20!2.3! Injury risk factors....................................................................................................... 20!2.3.1! Meeuwisse?s dynamic, recursive model of sport injury etiology ....................... 20!2.3.2! Non-modifiable risk factors ................................................................................ 21!2.3.2.1! Age group..................................................................................................... 21!2.3.2.2! Level of play and experience ....................................................................... 22!2.3.2.3! Position ........................................................................................................ 22!2.3.2.4! Previous injury ............................................................................................. 22!2.3.2.5! Sex................................................................................................................ 23!2.3.2.6! Anthropometrics .......................................................................................... 23!2.3.3! Modifiable risk factors........................................................................................ 24!2.3.3.1! Muscle strength............................................................................................ 24!2.3.3.2! Neuromuscular control and balance............................................................. 25!2.3.3.3! Flexibility..................................................................................................... 26!2.3.3.4! Aerobic fitness ............................................................................................. 26! v 2.3.3.5! Playing surface............................................................................................. 26!2.4! Injury prevention in youth soccer .............................................................................. 27!2.4.1! Injury prevention models .................................................................................... 27!2.4.2! Neuromuscular training for the prevention of injuries ....................................... 27!2.4.3! Limitations of previous research......................................................................... 38!2.5! Rationale, objectives and hypotheses ........................................................................ 38!2.5.1! Rationale ............................................................................................................. 38!2.5.2! Study 1: The effectiveness of a community-driven injury prevention program in reducing soccer-related injuries in female youth soccer players .................................... 40!2.5.3! Study 2: Risk factors for injury in youth female soccer players......................... 40!Chapter  3: Methods ............................................................................................................. 41!3.1! Purpose....................................................................................................................... 41!3.2! Research design ......................................................................................................... 41!3.3! Participants................................................................................................................. 41!3.3.1! Sample size ......................................................................................................... 41!3.3.2! Recruitment......................................................................................................... 41!3.3.3! Inclusion and exclusion criteria .......................................................................... 42!3.3.3.1! 2008-2009 .................................................................................................... 42!3.3.3.2! 2010-2011 .................................................................................................... 43!3.4! Data collection ........................................................................................................... 43!3.4.1! Baseline testing ................................................................................................... 43!3.4.1.1! Anthropometrics .......................................................................................... 44!3.4.1.2! Balance testing ............................................................................................. 44!3.4.1.3! Vertical drop jump ....................................................................................... 44!3.4.1.4! Aerobic fitness ............................................................................................. 45!3.4.1.5! Strength testing ............................................................................................ 46!3.4.2! Injury surveillance and documentation............................................................... 47!3.4.2.1! Operational definitions................................................................................. 48!3.4.2.2! 2008-2009 Injury surveillance ..................................................................... 48!3.4.2.3! 2010-2011 Injury surveillance ..................................................................... 49!3.5! Outcome measures ..................................................................................................... 49! vi 3.5.1! Primary outcome................................................................................................. 49!3.5.2! Secondary outcome............................................................................................. 50!3.6! Intervention ................................................................................................................ 50!3.6.1! Personnel and facilities ....................................................................................... 51!3.6.2! Injury prevention program .................................................................................. 52!3.7! Statistical analyses ..................................................................................................... 57!3.7.1! Predictive and descriptive statistics .................................................................... 57!Chapter  4: The effectiveness of a community-driven injury prevention program in reducing soccer-related injuries in female youth soccer players...................................... 58!4.1! Introduction................................................................................................................ 58!4.2! Methods...................................................................................................................... 59!4.2.1! Statistical analysis............................................................................................... 60!4.2.1.1! Descriptive epidemiology ............................................................................ 60!4.2.1.2! Risk ratio...................................................................................................... 60!4.3! Results........................................................................................................................ 61!4.3.1! Baseline characteristics....................................................................................... 61!4.3.2! Descriptive epidemiology ................................................................................... 65!4.3.2.1! Incidence proportion .................................................................................... 65!4.3.2.2! Injury location.............................................................................................. 66!4.3.2.3! Injury type.................................................................................................... 66!4.3.2.4! Injury severity .............................................................................................. 67!4.3.2.5! Injury mechanism and session type ............................................................. 68!4.3.3! The effectiveness of the SIPP ............................................................................. 69!4.3.3.1! Incidence proportion. Is the intervention effective at reducing injury?....... 69!4.3.3.2! Sensitivity analysis....................................................................................... 70!4.3.3.3! Effect modification by age group ................................................................ 70!4.3.3.4! Incidence proportion. Is the intervention effective at reducing injury after controlling for age group and level of play?............................................................... 71!4.3.3.5! Exploratory analysis..................................................................................... 72!4.4! Discussion .................................................................................................................. 73!4.4.1! Descriptive epidemiology ................................................................................... 73! vii 4.4.1.1! Incidence proportion .................................................................................... 73!4.4.1.2! Age group..................................................................................................... 75!4.4.1.3! Injury location.............................................................................................. 75!4.4.1.4! Injury type.................................................................................................... 76!4.4.1.5! Injury severity .............................................................................................. 77!4.4.1.6! Injury mechanism and session type ............................................................. 77!4.4.2! The effectiveness of the SIPP ............................................................................. 78!4.4.2.1! Effect of intervention ................................................................................... 78!4.4.2.2! Effect modification ...................................................................................... 78!4.4.2.3! Effect of intervention- controlling for covariates ........................................ 78!4.4.3! Implementation context ...................................................................................... 80!4.4.4! Uptake of the SIPP.............................................................................................. 81!4.4.5! Limitations .......................................................................................................... 81!4.4.5.1! Power ........................................................................................................... 81!4.4.5.2! Selection bias ............................................................................................... 81!4.4.5.3! Measurement bias ........................................................................................ 82!4.4.5.4! Confounding ................................................................................................ 83!4.4.5.5! Generalizability............................................................................................ 83!4.4.6! Strengths ............................................................................................................. 83!4.5! Summary and conclusions ......................................................................................... 84!Chapter  5: Risk factors for injury in youth female soccer players ................................. 85!5.1! Introduction................................................................................................................ 85!5.2! Methods...................................................................................................................... 86!5.2.1! Statistical analysis............................................................................................... 87!5.2.1.1! Descriptive epidemiology ............................................................................ 87!5.2.1.2! Incidence rate ratios ..................................................................................... 88!5.3! Results........................................................................................................................ 89!5.3.1! Baseline characteristics....................................................................................... 89!5.3.2! Descriptive epidemiology ................................................................................... 95!5.3.2.1! Injury rates ................................................................................................... 95!5.3.2.2! Injury location and injury type..................................................................... 97! viii 5.3.2.3! Injury severity .............................................................................................. 98!5.3.2.4! Injury mechanism and session type ............................................................. 98!5.3.3! Risk factors for injury ......................................................................................... 99!5.3.3.1! All injury...................................................................................................... 99!5.3.3.2! Lower-extremity injury.............................................................................. 100!5.4! Discussion ................................................................................................................ 105!5.4.1! Descriptive epidemiology ................................................................................. 105!5.4.1.1! Injury rates ................................................................................................. 105!5.4.1.2! Injury location and type ............................................................................. 106!5.4.1.3! Injury severity ............................................................................................ 107!5.4.1.4! Injury mechanism and session type ........................................................... 108!5.4.2! Risk factors for injury ....................................................................................... 108!5.4.2.1! Age group and level of play....................................................................... 109!5.4.2.2! Position ...................................................................................................... 109!5.4.2.3! Previous injury, sports participation, BMI, aerobic fitness, and balance .. 110!5.4.2.4! Vertical drop jump (VDJ) .......................................................................... 111!5.4.2.5! Strength ...................................................................................................... 111!5.4.3! Limitations ........................................................................................................ 112!5.4.3.1! Power ......................................................................................................... 112!5.4.3.2! Selection bias ............................................................................................. 113!5.4.3.3! Measurement bias ...................................................................................... 113!5.4.3.4! Confounding .............................................................................................. 114!5.4.3.5! Generalizability.......................................................................................... 114!5.4.4! Strengths ........................................................................................................... 114!5.5! Summary and conclusions ....................................................................................... 114!Chapter  6: Summary and conclusions ............................................................................. 115!6.1! Recommendations for implementation of the SIPP................................................. 116!References............................................................................................................................ 117!Appendices........................................................................................................................... 125!Appendix A Sample Size Calculations ............................................................................. 125!Appendix B Consent Form ............................................................................................... 126! ix Appendix C Preseason Questionnaire............................................................................... 129!Appendix D Physical Activity Readiness-Questionnaire ................................................. 131!Appendix E 2008-2009 Microsoft Access Injury Survey................................................. 133!Appendix F Weekly Exposure Sheet ................................................................................ 134!Appendix G Injury Report Form....................................................................................... 135!Appendix H Therapist Assessment Form ......................................................................... 138!Appendix I Physician Assessment Form .......................................................................... 139!Appendix J Injury Prevention Program Exercise Card..................................................... 140!Appendix K Soccer Injury Prevention Program Brochure ............................................... 142!  x List of tables Table 1 Prospective studies examining the incidence of injury in youth soccer ...................... 8!Table 2 Injury locations and types in prospective studies in youth soccer............................. 15!Table 3 Severity of injury by time loss categories (percent of all injuries)............................ 19!Table 4 Injury prevention studies in youth soccer .................................................................. 33!Table 5 Exercises and repetitions of the injury prevention program used as a structured warm-up program. See Appendix J for pictures. .................................................................... 54!Table 6 Exercises and repetitions for the home-based wobble board component of the injury prevention program. See Appendix J for pictures. ................................................................. 56!Table 7 Baseline characteristics by cohort.............................................................................. 62!Table 8 Baseline characteristics by cohort and injury status .................................................. 63!Table 9 Additional baseline characteristics for the 2010-2011 cohort ................................... 64!Table 10 Incidence proportion by cohort and age group ........................................................ 66!Table 11 Injury location and injury type by cohort ................................................................ 67!Table 12 Injury severity by cohort.......................................................................................... 68!Table 13 Injury mechanism and session type by cohort ......................................................... 69!Table 14 Effect of the SIPP on all injury, acute-onset injury and lower-extremity injury ..... 69!Table 15 Effect of the SIPP on all injury, acute-onset injury and lower-extremity injury with players that appear in both cohorts removed .......................................................................... 70!Table 16 Sub-group analysis of Risk Ratio by age group ...................................................... 71!Table 17 Risk factor analyses for all injury, acute-onset injury and lower-extremity injury . 72!Table 18 Exploratory analysis of specific injuries.................................................................. 73!Table 19 Baseline characteristics, obtained through preseason questionnaire and official roster information.................................................................................................................... 91!Table 20 Baseline characteristics, physical attributes............................................................. 93!Table 21 Baseline characteristics by injury status .................................................................. 94!Table 22 Incidence proportion and injury rate for all injury, acute-onset injury and lower-extremity injury....................................................................................................................... 96!Table 23 Injury location and injury type................................................................................. 97!Table 24 Injury severity .......................................................................................................... 98!Table 25 Injury mechanism and session type ......................................................................... 98! xi Table 26 Risk factor analyses for all injury .......................................................................... 101!Table 27 Risk factor analyses for lower-extremity injury .................................................... 103!   xii List of figures Figure 1 van Mechelen model of sports injury prevention research (39) ................................. 5!Figure 2 A dynamic, recursive model of etiology in sport injury (51) ................................... 21!Figure 3 The ?Translating Research into Injury Prevention Practice? framework (82) ......... 39!Figure 4 Vertical drop jump low-risk (left) and high-risk (right) landing.............................. 45!Figure 5 ?Soccer Injury Prevention Program? DVD .............................................................. 51!Figure 6 Flow chart describing recruitment of players into the study. ................................... 61!Figure 7 Number of players who participated in the various components of baseline testing90!   xiii List of abbreviations  Abbreviations Definition ABD Abduction ACL Anterior cruciate ligament ADD Adduction AE Athlete exposure ASIS Anterior superior iliac spine BC British Columbia BMI Body mass index CHIRPP Canadian Hospitals Injury Reporting and Prevention Program CI Confidence interval cm Centimetre CON Control EXT Extension F Female FIFA F?d?ration Internationale de Football Association FLX Flexion F-MARC FIFA Medical and Research Centre GP General practitioner HC High compliance tertile IC Intermediate compliance tertile INT Intervention IP Incidence proportion IRF Injury report form IR Injury rate IRR Incidence rate ratio LC Low compliance tertile LE Lower extremity IRR Incidence risk ratio kg Kilogram  xiv  Abbreviations Definition m Metre M Male min Minute mL Millilitre NEISS National Electronic Injury Surveillance System s Second SD Standard deviation OR Odds ratio PAR-Q Physical Activity Readiness Questionnaire PEP Prevent injury and Enhance Performance  PQ Preseason questionnaire RCT Randomized controlled trial RIO Reporting Information Online ROM Range of motion RR Risk ratio SD Safety director SIPP Soccer Injury Prevention Program TRIPP Translating Research into Injury Prevention Practice VDJ Vertical drop jump VO2 max Maximal aerobic power WES Weekly exposure sheet   xv Acknowledgements This thesis would not have been possible without the help of many people along the way. First, I would like to thank my supervisor Dr. Carolyn Emery. Your drive, knowledge and determination are unmatched. I have learned so much about sports injury prevention and research through your guidance and mentoring. I am extremely grateful for all of the opportunities that you have provided me throughout this degree. I am also grateful to Dr. Karim Khan for acting as my supervisor at UBC. Thank you for providing advice on research, constructive feedback on my study chapters and general ?life lessons?. I would also like to thank my committee member Dr. Shelina Babul who was incredibly helpful and supportive throughout my time at UBC. Thanks to Dr. Kristin Campbell for acting as my external examiner. Finally, I owe a special thanks to Dr. Jane Kang for her statistical guidance. I couldn?t have asked for a more knowledgeable, kind and generous person to learn statistics from. Thanks to all of you for being so generous with your time.  This work could not have been completed without Kerry Olohan and her incredible passion towards injury prevention in youth female soccer. I hope the findings from this study will continue to fuel your desire to prevent injuries in the community. I would also like to thank Marina Watson and Sam Goski for all of the time and hard work they put into the study. I also owe a big thank you to the Semiahmoo Soccer Club. Without the support of the players, coaches and parents, this project would not have been possible.   I would also like to acknowledge the Alberta Heritage Foundation and the Canadian Institutes of Health Research (Team in Child and Youth Injury Prevention) for their generous financial support.  Thank you to all of my officemates on the 6th floor at CHHM. It was a pleasure to work in such a supportive and cheerful environment. I would like to thank Christa Hoy for being such a great friend. Thank you for always providing me with support, encouragement and solutions to my problems. I owe a special thank you to Dr. Lindsay Nettlefold for answering my millions of questions and guiding me through the completion of my thesis. There is no way I could have accomplished this without your endless mentoring and support.   Finally, I would like to thank my family for all of their support during the last two years. On to the next one!  1 Chapter  1: Introduction  Soccer is one of the most popular sports in Canada with approximately 740,000 youth players registered in 2008.(1) While there are many benefits associated with physical activity, participation in sport comes with an inherent risk of injury. The rate of injury during practice is 1 to 5 injuries per 1,000 practice-hours for girls (2-7) and 2 to 6 injuries per 1,000 practice-hours for boys.(6,8-10) The overall rate of injury for game play is 8 to 22 injuries per 1,000 game-hours for girls (2-7) and 6 to 43 injuries per 1,000 game-hours for boys.(6,8-11) The majority of injuries sustained by youth soccer players occur to the lower-extremity (67-100%).(2-6,8-16) Injury in sport is associated with decreased levels of physical activity and long-term sequelae such as osteoarthritis.(17,18)   Although there is inconclusive evidence supporting differences in injury rates between girls and boys, there is evidence to suggest that the types of injuries suffered are different. Females demonstrate a 4 to 6-fold increased incidence of knee injury compared to males playing the same sports.(6,19-22)   Neuromuscular control has been identified as a modifiable risk factor for injury in youth soccer players, particularly in girls.(23,24) There is also evidence that previous injury (6,13,16,25), age group (5,6,13,16,26-29), skill level (6,30), joint laxity (31,32), and hamstring-to-quadriceps ratio (31) may be risk factors for injury in soccer. Injury prevention programs that target neuromuscular control may be a realistic option to reduce the impact of soccer-related injuries on girls.   Sports injury prevention is a growing area of interest and a number of studies have been carried out to assess the relationship between neuromuscular training and injury rates in youth soccer. Five studies have found a reduction in injury with use of a multifaceted injury prevention program in youth female soccer players (2,12,16,33,34) while one study found no difference in injury rates between the intervention and control groups.(3) These injury prevention programs reduced the risk of overall injury (2,12,16), anterior cruciate ligament  2 (ACL) injury (33), severe injury (2), acute-onset injury (16), and overuse injury (2) in youth female soccer players.   While there is clear support for the use of neuromuscular training to prevent injury in youth female soccer players, the evidence to date comes predominantly from efficacy studies, which focus on evaluation under ideal conditions (e.g., randomized controlled trials [RCT]). There is a paucity of effectiveness and implementation research that is needed to better understand how these interventions operate and succeed in the real-world. The community setting is the natural mode of delivering injury prevention programs and the focus of injury prevention research in youth soccer now needs to shift towards knowledge translation and how injury prevention programs can be best applied in a community setting.   Emery and Meeuwisse provide RCT support for neuromuscular training for the prevention of injuries in youth soccer.(16) The ?Soccer Injury Prevention Program? (SIPP) is designed to target strength, neuromuscular control and balance risk factors in order to prevent injury, particularly lower-extremity (LE) injury.(16) There is evidence that the SIPP decreased the risk of all injuries by 38% and acute-onset injuries by 43%.(16) The efficacy of the SIPP is understood but the effectiveness of this specific program when driven and delivered by a soccer community is unknown. This study provides a unique opportunity to evaluate a community-driven injury prevention program and to compare the results with the RCT efficacy results reported by Emery & Meeuwisse.(16)   The aim of this thesis is to determine the relationship between injury, risk factors and community-driven injury prevention in youth female soccer players. Chapter 2 provides relevant background information from which the research question, objectives and hypotheses are derived. Chapter 3 describes the study design and the methodology used to collect data in this study. Chapter 4 and 5 are written in paper format and discuss the findings related to each research question. Specifically, Chapter 4 addresses the question ?What is the effect of a community-driven neuromuscular training program on the incidence proportion of injuries of time loss ? 7 days in youth female soccer??. Chapter 5 addresses the question ?What are the risk factors for all injury and lower-extremity injury in youth female soccer  3 players participating in a neuromuscular training program??. Finally, Chapter 6 provides the summary and conclusion of findings as well as suggestions for future research in the field of youth female injury prevention in soccer.   4 Chapter  2: Literature review, objectives and hypotheses  2.1 Participation in youth soccer Soccer is the most popular sport played worldwide with an estimated 265 million players involved in the sport.(35) As participation in the sport continues to grow around the world, a similar trend has been seen in Canada. According to Stats Canada, 44% of Canadian children participated in soccer in 2005, making it the sport of choice among children ages 5 to 14.(36) Over 740,000 youth players were registered in Canada in 2008 of which 328,000 were female.(1) Participation in soccer in Canada has increased every year from 1980 to 2008 and female participation specifically, has increased every year from 1996 to 2008.(1) In British Columbia (BC) there were approximately 38,000 registered youth female soccer players during the 2011 to 2012 season.(37)  2.2  The epidemiology of soccer-related injuries Playing sports comes with an inherent risk of injury. Between 2003 and 2007, children and youth in BC experienced 5,644 injury hospitalizations from sports and recreation.(38) In children and youth between ages 10 to 19, soccer was the sixth leading cause of sports and recreation hospitalizations.(38) It has been reported that up to 8% of youth drop out of sport annually because of injury.(17) In addition to decreased participation in sport, there are also long-term sequelae associated with injury. The most common longer-term sequela associated with soccer-related injuries is osteoarthritis, a degeneration of cartilage within the joint. It has been reported that approximately 50% of patients with an anterior cruciate ligament (ACL) injury exhibit signs of osteoarthritis 10 years post-injury.(18)  With high participation rates across the country and within the province of BC, athletic injuries in soccer are an important public health concern. The first stage of van Mechelen?s sports injury prevention model is injury surveillance, the objective of which is to understand the burden of injury in the population of interest (Figure 1). Injury surveillance studies are necessary to provide information about the incidence, types and severity of injuries that occur in sport in order to inform injury prevention measures. The primary objective of this thesis  5 pertains to the final stage of van Mechelen?s four-stage model of injury prevention.(39) Stage four focuses on evaluating the effectiveness of injury prevention measures.               2.2.1 Injury definition Many studies use different definitions to define a sports injury, which makes comparisons of studies difficult. Examples of injury definitions include: time loss, medical attention, or tissue damage. Additionally, there are different data-collection methods that affect which injuries are captured and the accuracy of the data collected. The majority of soccer-related injury studies use an injury surveillance system. The different systems for capturing injuries prospectively have been administered by: athletic trainers, physicians, physiotherapists, coaches, medical students, players, specialists, referees, event medical staff or team designates. Other ways of capturing injury data include questionnaires, internet-based systems, and hospital-based injury reporting systems. Many descriptive epidemiologic studies include soccer-related injuries captured from existing hospital injury reporting systems such as the National Electronic Injury Surveillance System (NEISS) or the Canadian Hospitals Injury Reporting and Prevention Program (CHIRPP). This information is important 1. Establishing the extent of the sports injury problem ? Incidence ? Severity 2. Establishing etiology and mechanism of injuries 4. Assessing their effectiveness by repeating step 1 3. Introducing preventive measures Figure 1 van Mechelen model of sports injury prevention research (39)  6 in order to understand soccer-related injuries that require medical attention but these studies underestimate the true burden of injury in soccer.   Injury incidence is usually reported as either injury rates (# injuries/1,000 player-hours) or incidence proportion (# injuries/100 players). Rates may also be expressed as the number of injuries per 1,000 athlete exposures (AEs). An AE is equivalent to one game or practice and does not take into account the duration of the session. The number of injuries per 1,000 player-hours is considered the standard for reporting injury data since it provides the ability to account for different exposure risk between players. Presenting injury rates per 1,000 player-hours also allows comparison between different studies, sports and populations. In summary, differences in injury definitions and methodology have made it challenging to compare studies on soccer injuries. In 2006, a Consensus Statement was published to provide researchers with agreed upon injury definitions and data collection procedures for future studies on soccer injuries.(40)  2.2.2 Injury incidence The overall injury rate for youth soccer players in Calgary, Alberta is 3 to 6 injuries per 1,000 player-hours with no significant difference seen between indoor and outdoor soccer.(6,14,16) Some studies have found girls to be at greater risk of injury than boys (26,27,41) while other studies have found no difference in injury rates between girls and boys.(6,14,16) Yard et al. observed that girls had an increased risk of injury in games (incidence rate ratio [IRR], 1.25; 95% CI, 1.10-1.42) while boys had an increased risk of injury in practices (IRR, 1.24; 95% CI, 1.06-1.46).(42) The rate of injury during practices is 1 to 5 injuries per 1,000 practice-hours in girls (2-7) and 2 to 6 injuries per 1,000 practice-hours in boys.(6,8-10)   Game and tournament play consistently show higher rates of injury than practice sessions for both girls and boys. Emery et al. reported that the relative risk for injury in games compared to practice sessions was 3.26 (95% CI, 1.51-7.8) for boys and 2.57 (95% CI, 1.19-6.16) for girls.(14) The overall rate of injury for game play is 8 to 22 injuries per 1,000 player-hours for girls (2-7) and 6 to 43 injuries per 1,000 player-hours for boys.(6,8-11)   7  There are higher rates of injury during tournament play compared to regular season play. The overall youth injury rate for tournament play is 13 to 39 injuries per 1,000 player-hours.(43-45) Youth female players have a tournament injury rate between 14 to 51 injuries per 1,000 player-hours compared to 13 to 70 injuries per 1,000 player-hours for youth male soccer players.(43-45) In a 5-year study of tournament injuries, Williamson and Rice observed that boys had a significantly higher rate of injury compared to girls in the 12 to 15 year age group (69.9 injuries/1,000 hours versus 50.7 injuries/1,000 hours) and the U12 age group (25.9 injuries/1,000 hours versus 15.2 injuries/1,000 hours).(44) No difference was found between the tournament injury rates of boys and girls in the 16 to 18 year age group. Overall, boys were 1.24 times (95% CI, 1.01-1.52) more likely to sustain an injury during tournament play compared to girls.(44) The highest injury incidence rate, for both boys and girls, was in the 12 to 15 year age group.(44) A 10-year study of tournament injuries in the USA reported that girls had slightly higher rates of injuries than boys but that there was a trend towards this gap closing.(43) Table 1 summarizes prospective studies on the incidence of injury in youth soccer players.   8 Table 1 Prospective studies examining the incidence of injury in youth soccer Author  Study design (country, yr) Population (sex, age, level of play, other)  Sample size or denominator data Injury definition  Data-collection method No. of injuries Game injury rate, injuries/1,000 game-hours (95% CI)a   Practice injury rate, injuries/1,000 practice-hours (95% CI)a Overall injury rate, injuries/1,000 player-hours (95% CI)a Intervention studies      Heidt et al. (2000) (12)   Prospective cohort (USA) Female high school (ages 14-18)   n = 258 Time loss  Injury surveillance (athletic trainer) 91 (87 players) 15.1 injuries/100 players  20.2 injuries/100 players 35.3 injuries/100 players       Junge et al. (2002) (8)  Non-RCT (Switzerland, 1999-2000) Male amateur (ages 14-19; 3 high-level, 4 low-level teams)  n = 93 Physical complaint (> 2 weeks) or time loss  Injury surveillance (physician) 111 (67 players) 20.0  5.69 (includes overuse or practice injuries) 8.48        Mandelbaum et al. (2005) (33)  Prospective cohort (USA, 2000-2001) Female (ages 14-18, Coast Soccer League)  Year 1  n =1,905 Year 2 n =1,913 Confirmed noncontact ACL tears  Injury surveillance (coach); confirmation of noncontact ACL tear  Year 1  32 ACL tears Year 2  35 ACL tears   Year 1, 0.47 ACL injuries/1,000 AEs Year 2, 0.51 ACL injuries/1,000 AEs Overall, 0.49 ACL injuries/1,000 AEs    9 Author  Study design (country, yr) Population (sex, age, level of play, other)  Sample size or denominator data Injury definition  Data-collection method No. of injuries Game injury rate, injuries/1,000 game-hours (95% CI)a   Practice injury rate, injuries/1,000 practice-hours (95% CI)a Overall injury rate, injuries/1,000 player-hours (95% CI)a Steffen et al. (2008) (3)  Cluster RCT (Norway, 2005) Female (ages 13-17, U17 league)  n = 947 Time loss  Injury surveillance (physiotherapist) 241 (192 players) 7.6 (6.4-8.8)b  1.3 (1.0-1.6)b 3.7 (3.2-4.1)       Soligard et al. (2008) (2)  Cluster RCT (Norway, 2007) Female (ages 13-17, 15-16 year divisions)   n = 837 Time loss  Injury surveillance (coach, therapist, medical student) 215 9.6c  2.4c 4.7c       Emery & Meeuwisse (2010) (16)  Cluster RCT (Canada, 2006-2007) Male and female (ages 13-18, indoor soccer)  n = 364 Medical attention or time loss  Injury surveillance (designate, therapist) 79 (70 players)  3.35 (2.65-4.17)       Kiani et al. (2010) (46)  Non-RCT (Sweden, 2007) Female (ages 13-19)  n = 729 Acute knee injury (medical attention)  Injury surveillance (coach); hospital medical records 13 knee injuries 0.41 knee injuries/1,000 game-hours  0.08 knee injuries/1,000 practice-hours 0.14 acute knee injuries/1,000 player-hours       Wald?n et al. (2012) (47)  Cluster RCT (Sweden, 2009) Female (ages 12-17, Swedish Football Association)  n = 2,085 Acute knee injury (time loss)  Injury surveillance (physiotherapist, physician, coach) 47 knee injuries (44 players)  0.36 acute knee injuries/1,000 player-hours  10 Author   Study design (country, yr) Population (sex, age, level of play, other)  Sample size or denominator data Injury definition  Data-collection method No. of injuries Game injury rate, injuries/1,000 game-hours (95% CI)a  Practice injury rate, injuries/1,000 practice-hours (95% CI)a Overall injury rate, injuries/1,000 player-hours (95% CI)a Injury incidence studies      Females      Soderman et al. (2001) (4)  Prospective cohort (Sweden, 1996) Female (ages 14-20)  n = 153 Time loss  Injury surveillance system (trainer, coach, physiotherapist) 79 (63 players) 9.1b  1.5b  Overall, 6.8 Traumatic, 4.4       Le Gall et al. (2008) (5)  Prospective cohort (France, 1998-2006)  Female elite (ages 15-19, National Soccer Training and Development Centre for Females)   n = 119 Time loss (24 hours or more after the day of onset)  Injury surveillance system (physician) 619 (110 players) 22.4 (19.4-25.4)  4.6 (4.2-5.0)   Overall, 6.4 (5.9-6.9) Traumatic, 5.5 (5.0-6.0) Overuse, 0.9 (0.7-1.1) ACL, 1.0 (0.4-1.6)       Schiff et al. (2010) (7)  Prospective feasibility cohort (USA, 2006-2007) Female (ages 12-14, elite and recreational, Seattle Youth Soccer association)   n = 80 Medical attention or time loss  2 injury surveillance systems:  - parental internet-based system  - athletic therapist [reporting information online (RIO)] 44 10.6 (6.3-16.7)b  2.1 (1.0-4.0)b  Acute , 4.7 (3.1-6.8) Overuse, 2.9 (1.7-4.6)   11 Author   Study design (country, yr) Population (sex, age, level of play, other)  Sample size or denominator data Injury definition  Data-collection method No. of injuries Game injury rate, injuries/1,000 game-hours (95% CI)a  Practice injury rate, injuries/1,000 practice-hours (95% CI)a Overall injury rate, injuries/1,000 player-hours (95% CI)a Males and females      Kirkendall et al. (2002) (48)  Prospective cohort (USA) Male and female (ages U12-U18, Classic level)   n > 11,500 Time loss  Injury surveillance system (team report) 1,810 Boys, 10.89/1,000 AEs Girls, 9.81/1,000 AEs  Boys, 1.06/1,000 AEs Girls, 0.54/1,000 AEs Boys, 5.08/1,000 AEs Girls, 4.16/1,000 AEs       Emery et al. (2005) (6)  Descriptive epidemiology study (Canada, 2004) Male and female (ages 12-18, division 1-4)   n = 317 Medical attention or time loss  Injury surveillance system (designate, athletic therapist) 78 (61 players) Boys, 7.57 (5.11-10.78) Girls, 8.55 (5.78-12.18)  Boys, 2.94 (1.35-5.58) Girls, 2.62 (1.2-4.97) 5.59 (4.42-6.97)       Kucera et al. (2005) (13)  Prospective cohort (USA, 1997-2000) Male and female (ages 9-18, North Carolina Youth Soccer Association)   n = 1,483 Time loss  Injury surveillance system (coach) 905 (429 players)  4.6/1,000 AEs (4.3-4.9)       Yard et al. (2008) (42)  Descriptive epidemiology study (USA, 2005-2007) Male and female high school (nationally representative)  100 high schools 637,446 AEs Medical attention and time loss  Injury surveillance system (athletic therapist); RIO 1,524 Overall, 4.77/1,000 AEs Boys, 4.26/1,000 AEs Girls, 5.34/1,000 AEs  Overall, 1.37/1,000 AEs Boys, 1.51/1,000 AEs Girls, 1.21/1,000 AEs 2.39/1,000 AEs    12 Author   Study design (country, yr) Population (sex, age, level of play, other)  Sample size or denominator data Injury definition  Data-collection method No. of injuries Game injury rate, injuries/1,000 game-hours (95% CI)a  Practice injury rate, injuries/1,000 practice-hours (95% CI)a Overall injury rate, injuries/1,000 player-hours (95% CI)a Males      Peterson et al. (2000) (11)  Prospective cohort (Switzerland) Male adult and youth (ages 14-adult)   n = 264 Youth, n = 180 Tissue damage  Injury surveillance (physician) 558 (216 players) 16-18 years High-level, 18.9 Low-level, 42.5  14-16 years High-level, 15.8 Low-level, 37.8 16-18 years High-level, 6.6  Low-level, 13.7  14-16 years High-level, 6.0 Low-level, 11.4       Kakavelakis et al. (2003) (9)  Prospective cohort (Greece, 1999-2000) Male (ages 12-15)   n = 287 Time loss (limits participation for at least the day after the day of the onset)  Injury surveillance system (coach; specialist: orthopedic surgeon, general practitioner or pediatrician) 209 (193 players) 5.6  3.3  4.0       Brito et al. (2011) (10)  Prospective cohort (Portugal, 2009) Male sub-elite (ages 12-19)   n = 741  Time loss  Injury surveillance (physiotherapists, coaches) 53 6.8 (2.9-10.6)  1.8 (0.9-2.7)    2.5 (1.6-3.4)       Unknown      Le Gall et al. (2006) (15)  Prospective cohort  (France, 1993-2003)  Elite youth (U14-U16 age groups, National Institute of Football)   n = 528 Time loss (> 48 hours, not including the day of the injury)  Injury surveillance system (physician) 1,152 11.2  3.9  4.8   13 Author   Study design (country, yr) Population (sex, age, level of play, other)  Sample size or denominator data Injury definition  Data-collection method No. of injuries Game injury rate, injuries/1,000 game-hours (95% CI)a  Practice injury rate, injuries/1,000 practice-hours (95% CI)a Overall injury rate, injuries/1,000 player-hours (95% CI)a Tournament      Elias (2001) (43)  Prospective cohort (USA, 1988-1997) Male and female (International youth tournament, USA Cup)   n = 89,500 (estimated) Medical attention  Injury surveillance system (medical staff) 3,840  Boys, 12.69 (7.60-20.04)d Girls, 14.15 (10.23-20.11)d 13.23        Williamson & Rice (2006) (44)  Descriptive epidemiology study (2000-2004) Male and female (ages U8-U18, 5-day tournament)   52,140 player-hours Charted injuries seen by on-site sports medicine physicians  Injury surveillance system (physician) 367 Boys: U12, 25.9; ages 12-15, 69.9; ages 16-18, 31.7 Girls: U12, 15.2; ages 12-15, 50.7; ages 16-18, 20.7 35.7       Soligard et al. (2010)(45)  Prospective cohort (Norway, 2005-2008) Male and female (ages 13-19, Norway Cup tournament)   n > 60,000  Any injury, painful condition, or physical complaint sustained in a Norway Cup match, irrespective of the need for medical attention or time loss   Injury surveillance (coaches, referees) 2,454 Boys 38.1 Girls 42.0 39.2  aUnless stated otherwise; bOnly included acute injuries; cCalculated from available data; dRange (min-max);  Only control groups from the intervention studies are included in this table; CI, confidence interval; RCT, randomized controlled trial; ACL, anterior cruciate ligament; AE, athlete exposure; RIO, reporting information online  14 2.2.3 Injury location and type  The nature of soccer, as a fast-paced running and kicking sport with inevitable contact, lends itself to high rates of acute injuries primarily located in the lower-extremity. It has been reported that 67% to 100% of soccer related injuries are lower-extremity injuries.(2-6,8-16) Consistently, the most common injury locations in youth soccer are the knee (14-36%), ankle (16-35%), and thigh (6-25%).(2-6,8,9,11-13,15,42) Traumatic injuries account for the majority of outdoor soccer injuries with overuse injuries comprising only 9% to 37% of all injuries.(2-6,8,11,15,16) The most common types of injuries observed in youth soccer are sprains (17-47%), strains (15-25%), and contusions (9-31%).(2,3,5,11,15,27,42,49) This is true for both regular-season play and tournament play although there is evidence that contusions account for a greater proportion of injuries during tournament play.(44,45) Concussions account for between 0.3% and 6.4% of all injuries that occur in youth soccer.(4-6,9,10,44) Table 2 summarizes injury locations and injury types in prospective studies in youth soccer.  Although there is inconclusive evidence supporting differences in injury rates between girls and boys, there is evidence to suggest that the types of injuries suffered are different. The most common areas of the body injured in girls are the knee and ankle.(2-4,6) The rate of ACL tears for youth female soccer players has been reported to be 0.12 to 0.49 ACL tears/1,000 athlete hours.(5,33) Emery et al. reported that the top injury types for girls were ankle sprains, knee sprains and groin muscle strains compared to ankle sprains, concussions, groin muscle strains and calf muscle strains for boys.(6) Yard et al. estimated that girls were at a more than 3 times greater risk of complete ligament sprains compared to boys (risk ratio, 3.41; 95% CI, 1.39-8.39).(42) Furthermore, the rate of complete ligament sprains requiring surgeries was 13 times greater for girls compared to boys during competition.(42) Emery et al. reported a 5-fold greater risk of knee ligament sprain in girls compared to boys (IRR, 5.06; CI, 0.57-239.25).(6) In tournament play, girls experience lower rates of fractures and concussions compared to boys but higher rates of knee injuries and ankle sprains.(43)   15 Table 2 Injury locations and types in prospective studies in youth soccer  Intervention studies Injury incidence studies- females Study Heidt et al. (2000) (12) Junge et al. (2002) (8) Steffen et al. (2008) (3) Soligard et al. (2008) (2) Emery & Meeuwisse (2010) (16) Soderman et al. (2001) (4)  Le Gall et al. (2008) (5)  Schiff et al. (2010) (7) Design Prospective cohort Prospective cohort Cluster RCT Cluster RCT Cluster RCT Prospective cohort Prospective cohort Prospective feasibility cohort Country USA Switzerland Norway Norway Canada Sweden France USA Season  1999-2000 2005 2007 2006-2007 1996 1998-2006 2006-2007 Players 258 (F) 93 (M) 947 (F) 837 (F) 364 (F & M) 153 119 80 Injuries 91 111 241 215 79 79 619 44 Age range (years)  14-18 14-19 13-17 13-17 13-18 14-20 15-19 12-14 Lower-extremity 100%  82.4%a 73.5% 75.9% 89% 83.4%  Overuse  36.9% 12.9% 24.2% 8.9% 34% 13.4% 38.6% Injury Location (%)         Head/neck       0.4  Trunk/back      8.9 8.2  Upper-extremity   17.6a,b   2.5 5.6  Hip/groin 6.6  6.7a 4.2  6.3 11.6c 11a Thigh 14.3 21.6 13.3a 7.9  19.0 20.7  Knee 31.9 16.2 14.3a 27.0  19.0 16.8 11a Lower leg 15.4   10.2  12.7 5.0  Ankle 28.6 16.2 35.2a 24.2  22.8 25.4 44a Foot 2.2     8.9 6.1  Other 1.1        Injury type (%)         Sprain 39.6  40.2a 46.6a  31.6 26.9  Strain 24.2  20.1a 17.2a  19.0 25.2  Contusion 17.6  25.8a 20.2a  8.9 16.1  Concussion 0     2.5 0.3  Fracture 5.5   4.3a  2.5 3.2  Dislocation 1.1      0.3  Other 12.1  13.9a   35.4 28.0      16  Injury incidence studies- males and females Injury incidence studies- males Unknown Study Emery et al. (2005)d (6) Kucera et al. (2005)e (13) Emery & Meeuwisse (2006) (14) Yard et al. (2008) (42) Peterson et al. (2000)f (11) Kakavelakis et al. (2003) (9) Brito et al. (2011) (10) Le Gall et al. (2006) (15) Design Descriptive epidemiology Prospective cohort Prospective cohort Descriptive epidemiology Prospective cohort Prospective cohort Prospective cohort Prospective cohort Country Canada USA Canada USA  Greece Portugal France Season 2004 1997-2000 2004 2005-2007  1999-2000 2009 1993-2003 Players 317 1,483 142  180 287 741 528 Injuries 78 787 35 1,524 363 209 53 1,152 Age range (years)  12-18 9-18 13-17  14-adult 12-15 12-19 U14-U16 Lower-extremity 78.2% 66.9% 85.7% M, 73.0% F, 74.5% 79.6% 84% 79.2% 71% Overuse 10.3%    23%   17.2% Injury Location (%)         Head/neck 10.3   13.7 3.6 3 4 1.7 Trunk/back 6.4    5.9 5 11b 9.8 Upper-extremity 3.8    5.4 12 6 10.3 Hip/groin 10.3    7.3  8 7.1c Thigh 6.4   13.1 14.5 9 23 24.5 Knee 19.2 16.5  18.7 17.7 36 9 15.3 Lower leg 7.7    9.5 6 9 5.2 Ankle 28.2 26.4  23.4 20.4 29 13 17.8 Foot 7.7    10.0  17 8.2 Other     5.6    Injury type (%)         Sprain 34.6 35.9  26.8 54.0 33 15 16.7 Strain 24.4   17.9 33.0 23 25 15.3 Contusion  28.8  13.8 4.2 21g 25 30.6 Concussion 6.4   10.8  1 2  Fracture  12.3   3.1 8 4 5.9 Dislocation     2.5  2 0.9 Other 34.6    3.2 14 27         17  Tournament Study Elias (2001) (43) Williamson & Rice (2006) (44) Soligard et al. (2010) (45) Design Prospective cohort Descriptive epidemiology study Prospective cohort Country USA  Norway Season 1988-1997 2000-2004 2005-2008 Players 89,500 (F & M) 52,140 player-hours (F & M) > 60,000 (F & M) Injuries 3,840 367 2,454 Age range (years)  9-19 U8-U18  13-19 Lower-extremity 65.5%  69.8% Overuse  8.7%  Injury Location    Head/neck 13.6 11.7 12.6 Trunk/back 8.6   Upper-extremity 12.3 15.0  Hip/groin    Thigh 10.7 9.0  Knee 16.1 21.3 13.9 Lower leg 8.3 9.8  Ankle 18.5 16.9 20.5 Foot 7.3 7.9  Other    Injury type    Sprain  14.2 5.3 Strain   7.4 Contusion  30.8 39.4 Concussion  3.5  Fracture  10.9  Dislocation    Other    aOnly acute injuries were presented; bUpper body injuries; cIncludes pelvis injuries; dOnly lists injury  type for 51 injuries; eType of injury calculated as a percentage of those injuries that had type listed;  fPercentages for injury location and injury type are for adult and youth injuries combined; gIncludes  abrasion injuries; Face injuries included in head/neck category; RCT, randomized controlled trial; F, female; M, male   18 2.2.4 Injury severity Table 3 provides a breakdown of the severity of injuries for prospective studies in youth soccer players. Time loss is presented as the number of days absent from soccer (practices or games) due to injury. Control group data is only provided for the intervention studies. It is evident that differences in definitions of injury and severity classifications make it difficult to make valid comparisons between categories of severity. For example, 6% to 37% of injuries in youth soccer are classified as severe where the definition for a severe injury ranges from ?greater than 14 days time loss from soccer? to ?season-ending? injuries.(2-6,8-11,15,16,42) Two prospective cohort studies in youth female soccer players have shown that 34% to 52% of injuries were minor (1-6 days absent), 36% to 52% were moderate (7-30 days absent) and 11% to 14% were classified as severe (> 30 days absent).(4,5)     19 Table 3 Severity of injury by time loss categories (percent of all injuries)  Study Population Severity classification (days absent)   Mild Minor Moderate Severe   (1-7)  (8-27) (>27) Junge et al. (2002) (8) Male amateur (ages 14-19; 3 high-level, 4 low-level teams) 66.7  19.8 13.5    (1-7) (8-21) (>21) Steffen et al. (2008) (3) Female (ages 13-17, U17 league)  39.0a 32.9a 28.1a   (1-3) (4-7) (8-28) (> 28) Soligard et al. (2008) (2) Female (ages 13-17, 15-16 year divisions) 14.9 15.8 32.6 36.7   (0-7)  (8-28) (> 28) Emery & Meeuwisse (2010) (16) Male and female (ages 13-18, indoor soccer) 77.2  16.5 6.3    (1-6) (7-30) (>30) Soderman et al. (2001) (4) Female (ages 14-20)  34.0 52.0 14.0    (1-7) (7-30) (>30) Le Gall et al. (2008) (5)  Female elite (ages 15-19, National Soccer Training and Development Centre for Females)  51.9 35.7 10.7   (0-1) (2-7) (8-14) (>14) Emery et al. (2005) (6) Male and female (ages 12-18, division 1-4) 43.6 30.8 10.3 15.4   (<7)  (>21) (Season-ending) Yard et al. (2008) (42) Male and female high school (nationally representative) 55.8  5.8 8.7   (0-7)  (8-27) (>27) Peterson et al. (2000) (11) Male adult and youth (ages 14-adult)  52.2  32.4 15.4    (<7) (7-30) (>30) Kakavelakis et al. (2003) (9) Male (ages 12-15)   30.0 38.0 32.0   (1-3) (4-7) (8-28) (>28) Brito et al. (2011) (10) Male sub-elite (ages 12-19)  34.0 11.0 40.0 15.0   (1-3) (4-7) (7-28) (>28) Le Gall et al. (2006) (15)  Elite youth (U14-U16 age groups, National Institute of Football) 31.0 29.3 29.9 9.9 aOnly acute injuries were presented  20 2.2.5 Injury mechanism  Contact is a common mechanism of injury in soccer. Contact definitions may include contact between players, equipment or the ground depending on the study. Player-to-player contact was the mechanism of injury for 36% to 43% of injuries that occurred in youth soccer players.(8,12,42,49) Similarly, Emery et al. reported that 46.2% (95% CI, 34.8-57.8%) of all injuries were a result of direct contact with another player or equipment.(6)   2.3 Injury risk factors  2.3.1 Meeuwisse?s dynamic, recursive model of sport injury etiology Building on previous work (50), Meeuwisse et al. proposed a dynamic model of etiology in sport injury (Figure 2).(51) The original model of etiology in sport injury was linear while the new model accounts for the dynamic and fluid nature of injury risk. Each athlete has an individual set of intrinsic risk factors (e.g., previous injury, sex, age, flexibility, strength, and neuromuscular control) that predispose the athlete to injury. A ?predisposed athlete? is exposed to extrinsic risk factors (e.g., playing surface, rules, level of play, and equipment) and the combination of intrinsic and extrinsic risk factors produces the ?susceptible athlete.?(51) Events can lead to injury or more commonly, no injury. Non-injury events may bring about adaptation and modification of the risk factors acting on the athlete. Injury events may lead to recovery (with possible adaptation and modification of risk factors) or no recovery and subsequent removal from sport. This model represents the true cyclical nature of varying risk profiles, causation, repeated exposure and injury.      21   Figure 2 A dynamic, recursive model of etiology in sport injury (51)  2.3.2 Non-modifiable risk factors Non-modifiable risk factors are those that cannot be changed or eliminated. Although these risk factors cannot be altered, they are important to sport injury research in order to tailor injury prevention interventions towards athletes that are most at risk.  2.3.2.1 Age group The majority of previous studies in youth soccer have found that injury risk is greater in older age groups compared to younger age groups.(13,16,26-29) In contrast, two studies found higher rates of injury in younger age groups compared to older age groups.(5,6) These conflicting results may be explained by physiological differences between age groups or underreporting of injuries in the older age groups.(6) Older age was reported to be a significant risk factor for injury in two adult soccer studies.(32,52) Other studies in adult soccer have found no evidence of age being a significant predictor of overall injury risk in  22 males (53) or females.(31,54) There is evidence that age is a significant risk factor for hamstring injury in both elite (53) and professional (55) male soccer players.   2.3.2.2 Level of play and experience Results in the literature are mixed with some studies reporting an increased risk of injury in the more elite divisions of play (6) or more skilled players (30), and other studies reporting an increased risk of injury in the lower divisions of play.(11,56) Kucera et al. found that players with > 2 years experience had a reduced risk of injury compared to those players with ? 2 years experience.(13) The protective association between increasing soccer experience and injury may be explained through a ?survival effect? where players that do not experience an injury are more likely to continue participation in the sport than those players that experience an injury.(13)   2.3.2.3 Position Results in the literature are inconsistent regarding position of play as a risk factor for injury and little research has been done in youth populations. Kucera et al. found that defenders were at significantly greater risk for injury than midfield players in youth soccer.(13) In a study in women?s professional soccer, players in the midfield position sustained a significantly higher number of injuries compared to players in all other positions.(57) Another study in elite women?s soccer showed that there was a higher incidence rate in defenders and forwards compared to goalkeepers and midfielders.(54) Furthermore, Le Gall et al. (5) and Jacobson & Tegner (58) found that there were no significant differences in injury incidence rates between the various positions in youth and adult female players respectively.   2.3.2.4 Previous injury Consistently, previous injury has been shown to be a risk factor for future injury in youth soccer players.(6,13,16,25) Emery et al. observed an increased risk of injury in youth soccer players that reported an injury in the previous year compared to those that did not (RR, 1.74; 95% CI, 1.0-3.1).(6) Additionally, there is evidence that previous injury is a risk factor for specific injury types including ankle (13,25,52,54,59), knee (13,25,52-54,60), hamstring  23 (52,53,61), and groin injuries (25,52,53,62) in soccer. Kucera et al. reported that youth soccer players with a history of ankle and knee injury were at a four and six times greater risk of ankle and knee injury respectively (ankle IRR, 4.19; 95% CI, 2.97-5.92; knee IRR, 5.84; 95% CI, 4.04-8.44).(13)   Frisch et al. did not find an association between previous injury and injury risk in youth male soccer players.(63) This study was limited by the small sample size (n = 67). Similarly, Soderman et al. found no evidence of a relationship between previous injury and leg injuries in adult female soccer players, although previous injury was limited to the three months prior to the start of the study.(31) Contrary to the majority of the literature, a study in 100 professional soccer players in Greece reported that players with a previous hamstring strain were at decreased odds of injury (OR, 0.15; 95% CI, 0.029-0.79).(64)   Overall, the literature supports previous injury as a risk factor for incident injury in soccer. Previous injury is likely a risk factor for incident injury due to physiological deficiencies, anatomical deficiencies or inadequate rehabilitation following the previous injury.(6)  2.3.2.5 Sex Sex differences were previously discussed in the sections on injury incidence (2.2.2), injury location and injury type (2.2.3). In summary, there is some evidence that girls have a greater risk of injury than boys (26,27,41) but more recent work has found that there is no difference in the rates of injury between girls and boys.(6,13,14,16) There is evidence that the types of injuries sustained by each sex are different, with girls sustaining more sprains than boys.(6,42,43,65)   2.3.2.6 Anthropometrics  Although weight may be considered a modifiable risk factor, anthropometrics have been classified as non-modifiable risk factors since height is non-modifiable and these risk factors are often evaluated together. A study in German elite women?s soccer, found evidence that taller players (> 1 SD above the mean) were at increased risk for injury compared with those that were intermediate in height.(54) Using a univariate approach, Arnason et al. observed  24 that increased body fat was a risk factor for groin strain injury in male soccer players.(52) However, when using multivariate logistic regression analysis, body fat was no longer found to be a significant predictor of groin strain injury.   Often weight and height are evaluated together using body mass index (BMI = weight/height2). Kucera et al. reported that increasing BMI (using BMI quartiles) was associated with an increased risk of injury.(13) However, once adjusted for other covariates this relationship was no longer found to be significant. Furthermore, studies in adult male (52), adult female (32), and youth (6) soccer players have also found no association between BMI and injury risk.   2.3.3 Modifiable risk factors Modifiable risk factors are those risk factors that can be changed. The objective of injury prevention programs is to target modifiable risk factors in order to reduce injury rates in the population of interest.  2.3.3.1 Muscle strength Knee flexor and extensor strength have been evaluated as potential risk factors for injury in soccer. There is evidence that females with a lower hamstring-to-quadricep ratio (concentric) are at increased risk of injury in soccer.(31,66) On the other hand, a higher ratio was found to be a significant risk factor for overuse injury.(31) A study in 100 professional male soccer players found that eccentric isokinetic strength asymmetry of > 15% increased the risk of noncontact hamstring strain.(64) Orchard et al. demonstrated an increased risk of hamstring injury in Australian Rules Football players with lower flexion-to-extension ratios obtained during an isokinetic strength test at 60?/s.(67) Finally, ?stenberg & Roos did not find an association between isokinetic muscle strength and risk of injury in adult female soccer players.(32) Muscle strength is modifiable and strengthening exercises are easily incorporated into injury prevention programs. As a result, strengthening exercises are a common component of all injury prevention programs in youth soccer.(2,3,8,12,16,33,46,47)    25 2.3.3.2 Neuromuscular control and balance There appears to be differences in the kinematics around the knee and ankle between males and females.(24) During an unanticipated cut maneuver, females showed increased knee abduction angles and greater maximum ankle eversion (in stance phase) compared to males.(24) Differences in knee joint stability and neuromuscular control have been proposed as possible factors for the higher risk of ACL injury in females compared to males. Female athletes with increased dynamic knee valgus and abduction moment have been shown to be at increased risk of ACL injury.(23) Decreased maximal knee flexion has also been shown to be a risk factor for ACL injury in female athletes.(23)  Neuromuscular training programs have been primarily targeted towards females due to the gender differences observed. Plyometric training has been shown to decrease landing forces, adduction moments, abduction moments and increase knee stabilization in female youth volleyball players.(68) Furthermore, Myer et al. found that both plyometric and dynamic balance training may reduce lower-extremity knee valgus measures.(69) Plyometric training was shown to affect sagittal plane kinematics during a vertical drop jump test while balance training affected the sagittal plane kinematics during the single-leg drop landing test.(69) Since plyometric and balance training appear to affect knee stabilization during different tasks, it is proposed that both types of training should be included in injury prevention programs for maximal benefit.    Soderman et al. found that low postural sway of the legs was a significant risk factor for traumatic leg injury, where low postural sway indicates good balance (OR, 0.31; p = 0.005). Emery et al. did not find an association between dynamic balance and an increased risk of injury in youth soccer players.(6) A study including 508 amateur male soccer players found that there was no evidence that balance (tested on both the floor and a balance mat) was a significant predictor of ankle injury.(59) Additionally, Frisch et al. reported that static balance and dynamic balance were not significant risk factors for injury in youth soccer.(63) Although there is inconclusive evidence supporting balance as a risk factor for injury, there is evidence that participation in an injury prevention program that includes a balance component reduces the risk of injury in soccer.(2,16,46,70,71)  26 2.3.3.3 Flexibility General joint laxity was found to be a significant predictor of all injury (32), and lower-extremity injury (31) in adult female soccer players. Soderman et al. reported that there was an increased risk of leg injury in adult female soccer players that showed hyperextension of the knee joint > 10? (OR, 3.84; 95% CI, 1.51-9.78).(31) Furthermore, there was evidence that asymmetries in ankle dorsiflexion (OR, 7.06; p = 0.02) and hamstring flexibility (OR, 3.56; p = 0.049) were significant risk factors for overuse leg injuries.(31) Myer et al. also reported that knee hyperextension (beyond neutral) was a significant predictor of ACL injury in female athletes (OR, 4.78; 95% CI, 1.24-18.44).(72) In this study, side-to-side difference in knee laxity was also found to be a significant risk factor for ACL injury (OR, 4.03; 95% CI, 1.68-9.69). Henderson et al. observed that reduced active hip flexion range of motion (ROM) was a significant risk factor for hamstring injury in professional male soccer players.(55) Using multivariate logistic regression analysis, decreased ROM in hip abduction was found to be a risk factor for groin strain in male soccer players (OR, 0.9; 95% CI, 0.8-1.0).(52)  2.3.3.4 Aerobic fitness Previous studies in youth and adult groups have reported no association between aerobic fitness and injury risk in soccer.(6,32,52,63,73) These findings may be a result of the homogeneity of the study populations.   2.3.3.5 Playing surface Studies have found that there is no difference in injury rates between play on artificial turf and natural grass in youth (45,74,75) or adult soccer.(76-79) Although there is no evidence for playing surface being a risk factor for injury, there is some evidence that the types of injuries sustained differ between playing surfaces. Aoki et al. found that there was an increased incidence of low back pain in the artificial turf group compared to the natural grass group (IRR, 1.62; 95% CI, 1.06-2.48).(75) Steffen et al. reported an increased incidence of sprains and knee injuries in youth female soccer players on artificial turf compared to grass.(74) During tournament play, there was a decreased risk of ankle injuries and an increased risk of back/spine, and shoulder injuries on artificial turf compared to grass.(45) Ekstrand et al. reported that there was an increased risk of ankle sprain (IRR, 1.81; 95% CI,  27 1.00-3.28) and a decreased risk of strains (IRR, 0.60; 95% CI, 0.37-0.99) during game play on turf compared to grass.(77) Similarly, Fuller et al. reported an increased risk of ligament injuries, and foot and ankle injuries for males practicing on turf compared to grass.(78) Males were at an increased risk of sustaining a calf strain while practicing on grass compared to turf.(76)   2.4 Injury prevention in youth soccer 2.4.1 Injury prevention models The last stage of van Mechelen?s sports injury prevention model is the evaluation of the effectiveness of the injury prevention measure (Figure 1).(39) Using this model, the final stage of sports injury research involves efficacy research and determining the effects of the preventive measures under ?ideal conditions? (e.g., randomized controlled trials [RCT]).    2.4.2 Neuromuscular training for the prevention of injuries  Neuromuscular control refers to the ability to efficiently coordinate the joint systems in the body in order to adapt to the dynamic changes that occur during movement.(80) Neuromuscular training involves exercises that target the passive structures, active structures and the nervous system that is responsible for stability and mobility of the joints in order to improve neuromuscular control and prevent injuries. It has been hypothesized that increasing lower body strength and proprioceptive awareness can result in decreased injury rates among athletes. Hewett et al. conducted a study assessing the effect of a plyometric training program on jump and landing lower-extremity mechanics.(68) It was proposed that peak landing forces and knee abduction and adduction moments decreased among female volleyball players following 6 weeks of training.(68) This suggests that a multifaceted program focused on jumping and landing technique and improving strength may result in increased dynamic knee stability and subsequently a decrease in knee injuries in female athletes. Several neuromuscular training programs have been developed that aim to decrease injuries in an array of sports and a variety of populations.   28 As of May 2012, nine prospective studies have been published that have assessed the effectiveness of injury-prevention programs in youth soccer players.(2,3,8,12,16,33,34,46,47) The interventions used in these studies were multifaceted injury prevention programs. Of the nine intervention studies, seven included females only(2,3,12,33,34,46,47), one included males only (8), and one included both females and males.(16) Table 4 summarizes injury prevention studies in youth soccer.   Heidt et al. conducted a quasi-experimental trial to evaluate the effect of a preseason conditioning program on the incidence and severity of injuries in female high school soccer players.(12) The Frappier Acceleration Training Program included aerobic, plyometric, sport cord, strength and flexibility components.(12) Additionally, the program had an educational component that taught the athletes about proper movement technique and avoidance of high-risk movements.(12) The proportion of injured athletes was reported to be significantly higher in the untrained group compared with the trained group (33.7% versus 14.3%; p = 0.0085).(12) It was suggested that there was a greater proportion of ACL tears in the untrained group compared to the trained group (3.1% versus 2.4%).(12) This relationship did not reach statistical significance but is an important clinical finding.     Junge et al. evaluated the effect of an injury prevention program on youth male soccer players using a quasi-experimental study design.(8) This program-included interventions directed towards warm-up, cool-down, prophylaxes of unstable ankles, sufficient rehabilitation of injuries and promotion of fair play.(8) Of interest, the players participated in FIFA Medical and Research Center (F-MARC) Bricks, a series of ten exercises that included elements of stability, flexibility, strength, coordination, reaction time and endurance.(8) The injury rate was 6.7 injuries/1,000 player-hours for the intervention group and 8.5 injuries/1,000 athlete hours for the control group.(8) It was suggested that the injury rate was lower in the intervention group compared to the control group but the results were not statistically significant [IRR estimate, 0.79 (not significant)].(8) Skill level appeared to be an effect-modifier in this study. Low-skill teams seemed to gain greater benefits from the injury prevention program compared to high-skill teams.(8)   29 Mandelbaum et al. evaluated the effectiveness of the ?Prevent injury and Enhance Performance? (PEP) program on the incidence of ACL injuries.(33) Confirmed ACL tears were the primary outcome of interest in this study.(33) The PEP program includes warm-up, stretching, strengthening and plyometric exercises as well as soccer-specific agility drills.(33) In the first year of the study, there was an 88% reduction in ACL injuries in the intervention athletes compared to the control athletes.(33) In the second year of the study, a 74% reduction in ACL injuries was observed in the intervention group when compared to the control group.(33) The overall incidence rate for the trained group was 0.09 ACL tears/1,000 athlete exposures compared to 0.49 ACL tears/1,000 athlete exposures in the untrained group. During the two seasons of the study, the intervention was shown to reduce ACL tears by 82% (RR, 0.18; p < 0.0001).   In 2003, the F-MARC developed the ?11? injury prevention program. The ?11? combined elements from the PEP program and the F-MARC bricks, two injury prevention programs that suggested a reduction in the incidence of injuries in youth soccer players.(8,33) The ?11? consisted of 10 exercises that focus on core stability, balance, plyometrics and strength.(3) The eleventh component of the program is promotion of fair play but was not included in the study.   Steffen et al. conducted a cluster-RCT to evaluate the effectiveness of the FIFA ?11? injury prevention program.(3) There was no difference found in the proportion of injured players or in the injury incidence rate between the intervention and control groups.(3) A limitation of the study was that compliance to the intervention was low (52% of training sessions) and may have been inadequate to reduce the risk of injury.(3)   The findings of the previously mentioned study led to the development of the ?11+? injury prevention program.(3) The ?11+? is a revised version of the ?11? that provides variation and progression of the exercises, plus a new set of aerobic exercises so that the ?11+? program can be used in place of a standard warm-up.(2) Changes to the ?11? were made to improve the effectiveness of the program as well as to increase compliance to the intervention. Soligard et al. conducted a cluster-RCT to assess the effectiveness of the new ?11+? program  30 in reducing injuries in youth female soccer players in Norway.(2) There was a significantly lower risk of overall injury (IRR, 0.68; 95% CI, 0.48-0.98), overuse injury (IRR, 0.47; 95% CI, 0.26-0.85) and severe injury (IRR, 0.55; 95% CI, 0.36-0.83) in the intervention group when compared to the control group.(2) It was estimated that use of the ?11+? can reduce the risk of injury by 32% and severe injuries by 45%.(2) Compliance was higher in the ?11+? study than in the ?11? study (77% versus 52%).(2)  Soligard et al. did further analysis of the intervention group to evaluate the dose-response relationship of compliance on risk reduction.(34) A secondary objective of the study was to examine whether there was an association between coaches? attitudes toward injury prevention with compliance to the program and risk of injury. Compliance, injury and exposure hours were collected prospectively while coaches? attitudes towards injury prevention were collected retrospectively at the end of the intervention season.(34) It was found that team compliance to the ?11+? was 77% while player compliance was 79%.(34)   To evaluate the risk of injury associated with compliance, teams and players were stratified by tertile into high, intermediate and low compliance. There was no difference found in the incidence rate ratio between the high, intermediate and low compliance teams. When analyzed on an individual level, it was found that the risk of injury for players in the high compliance tertile was 35% lower than players in the intermediate tertile (IRR, 0.65; 95% CI, 0.46-0.91).(34) In addition, the risk of acute injury was 39% lower for players in the high compliance tertile when compared to the players in the intermediate tertile (IRR, 0.61; 95% CI, 0.42-0.88).(34) There was no significant difference found in the overall risk of injury or acute injury between the high compliance and the low compliance tertiles.   It was also reported that coach attitudes had an effect on team compliance. The probability of having low compliance with the program was 87% higher if the coach believed that the program took too much time (OR, 0.13; 95% CI, 0.03-0.60).(34) Furthermore, the probability of having low compliance with the program was 81% higher if the coach believed that the program did not include enough soccer-specific activities (OR, 0.19; 95% CI, 0.40-0.92).(34)  31 Coaches with positive attitudes towards injury prevention training were associated with teams that had high compliance to the injury prevention program.   Emery & Meeuwisse used a validated injury surveillance system (6) to assess the effectiveness of the ?Soccer Injury Prevention Program? (SIPP) in reducing injuries in youth soccer players.(16) This intervention included a neuromuscular training warm-up program and a home-based program. The control group was taught a ?standard of practice? warm-up that included aerobic, static stretching and dynamic stretching components.(16) In addition, the control group was given a stretching program for the home-based component of the training program.(16) The intervention group was taught a warm-up that included the same aerobic and dynamic stretching components as the control group but also included neuromuscular training exercises (strength, balance and agility).(16) The home-based component, for the intervention group, was a 15-minute wobble-board balance-training program that was supposed to be performed three times per week.(16) The adjusted incidence rate ratio (IRR) for all injury (IRR, 0.62; 95% CI, 0.39-0.99) and acute-onset injury (IRR, 0.57; 95% CI, 0.35-0.91) were found to be significant.(16) The SIPP was effective at decreasing the risk of all injuries by 38% and acute-onset injuries by 43%.(16) In addition, there was a trend of reduction in the risk of lower-extremity, ankle sprain and knee sprain injuries that were not statistically significant but should be considered clinically relevant.(16)   Two recent studies assessed the effectiveness of injury prevention programs at reducing the risk of acute knee injuries only.(46,47) Kiani et al. conducted a non-RCT in Sweden using the HarmoKnee preventive program.(46)The HarmoKnee program is designed to replace a standard warm-up and includes aerobic, muscle activation, balance, strength and core components. There was a significantly lower risk of knee injury in the intervention group compared to the control group (unadjusted RR, 0.23; 95% CI, 0.04-0.83).(46) Likewise, participation in the HarmoKnee preventive program reduced the risk of noncontact knee injury by 90% (unadjusted RR, 0.10; 95% CI, 0.00-0.70).(46) Major limitations of this study include inclusion of non soccer-related injuries and the small number of injuries that occurred over the study period.   32 Finally, the most recent injury prevention study in youth soccer evaluated the effectiveness of a neuromuscular training program (Kn?kontroll) in reducing the rate of acute knee injury, particularly ACL injury.(47) This intervention was designed to replace a standard warm-up and included 6 exercises with each exercise having levels of progressive difficulty. The unadjusted Cox regression showed that there was a significantly lower risk of ACL injury in the intervention group compared to the control group (RR, 0.36; 95% CI, 0.15-0.85). In addition, compliant players (those that participated in the intervention at least once/week) had a lower risk of severe knee injury (RR, 0.18; 95% CI, 0.07-0.45) and acute knee injury (RR, 0.53; 95% CI, 0.30-0.94) than noncompliant players.(47) There was no difference in acute knee injury risk between the intervention and control group (RR, 0.92; 95% CI, 0.61-1.40). This is likely due to the small number of injuries that were sustained over the study period.    33 Table 4 Injury prevention studies in youth soccer Author Study design Country Year Observation period Population (sex, age, level of play, other)  No. of players/no. of teams Injury definition  Targeted injuries Intervention Injury rates (injuries/1,000 player-hours)a   Effect of intervention Heidt et al. (2000) (12)  Prospective cohort   USA  1 year Female high school (ages 14-18)   Intervention [INT] (42)  Control [CON] (258). Time loss  All injury Frappier Acceleration Training Program - Sport-specific cardiovascular training, plyometric drills, sport cord drills, strength training and flexibility exercises  - Recognition and avoidance of high-risk movements  - 20 sessions over 7-weeks Overall INT, 16.7 injuries/100 players CON, 35.3 injuries/100 players Game INT, 11.9 injuries/100 players CON, 15.1 injuries/100 players Practice INT 4.8 injuries/100 players CON, 20.2 injuries/100 players  - There was a significantly higher incidence of injury in the untrained group when compared to the trained group (RR, 0.42; 95% CI, 0.20-0.91)       Junge et al. (2002) (8)  Quasi-experimental  Switzerland  1999-2000  2 seasons Male amateur (ages 14-19; high and low level teams)   INT (101/7)  CON (93/7) Physical complaint that lasted longer than two weeks or time loss  All injury - Improvement of warm-up, regular cooldown, taping of unstable ankles, adequate rehab, promotion of fair play  F-MARC Bricks - 10 exercises designed to improve joint stability, muscular strength, flexibility, coordination, reaction time and endurance Overall INT, 6.71 CON, 8.48 Game INT, 15.9 CON, 20.0 Practiceb  INT, 3.96 CON, 5.69   - There were fewer injuries in the intervention group compared to the control group [IRR estimate, 0.79 (not significant)] - The intervention seemed to be more beneficial at preventing injuries in low skill teams compared to high skill teams  34 Author Study design Country Year Observation period Population (sex, age, level of play, other)  No. of players/no. of teams Injury definition  Targeted injuries Intervention Injury rates (injuries/1,000 player-hours)a   Effect of intervention Mandelbaum et al. (2005) (33)  Prospective cohort   USA  2000-2001  2 years Female (ages 14-18, Coast Soccer League)   Year 1 INT (1,041/52) CON (1,905/95) Year 2 INT (844/45) CON (1,913/112) Confirmed noncontact ACL tears  Noncontact ACL injury Prevent injury and Enhance Performance program  - 19 exercises including warm-up, flexibility, strength, plyometric and agility components INT, 0.09/1,000 AEs CON, 0.49/1,000 AEs  - IRR, 0.181 (P < 0.0001) - Year 1, 88% reduction in ACL injury for the intervention group compared to the control group - Year 2, 74% reduction in ACL tears.      Steffen et al. (2008) (3)  Cluster RCT   Norway  2005  8 months Female soccer (ages 13-17, U17 league)   INT (1,073/58)  CON (947/51) Time loss  All injury FIFA ?11?  - 10 exercises including core stability, balance, plyometric and strength components   - Every training session for 15 consecutive sessions and then once a week for the remainder of the season Overall  INT, 3.6 (95% CI, 3.2-4.1) CON, 3.7 (95% CI, 3.2-4.1) Game INT, 8.2 (95% CI, 6.9-9.4)c CON, 7.6 (95% CI, 6.4-8.8)c Practice INT, 0.9 (95% CI, 0.6-1.2)c CON, 1.3 (95% CI, 1.0-1.6)c  - IRR, 1.0 (95% CI, 0.8-1.2) - No difference observed in the proportion of injured players, overall injury incidence or injury rates between intervention and control teams   35 Author Study design Country Year Observation period Population (sex, age, level of play, other)  No. of players/no. of teams Injury definition  Targeted injuries Intervention Injury rates (injuries/1,000 player-hours)a   Effect of intervention Soligard et al. (2008) (2)  Cluster RCT   Norway  2007  8 months Female (ages 13-17, 15-16 year divisions)   INT (1,055/52)  CON (837/41)  Time loss  All injury FIFA ?11+? - Revised version of the ?11? - Running, strength, balance, jumping and speed running components  - Includes additional exercises designed to add variation and progression  - Complete program used every training session while the running exercises in the program were used as part of the warm-up before every game Overall INT, 3.2 CON, 4.7 Game INT, 6.8 CON, 9.6 Practice INT, 1.5 CON, 2.4  - IRR all injuries, 0.68 (95% CI, 0.48-0.98) - IRR overuse injuries, 0.47 (95% CI, 0.26-0.85) - IRR severe injuries, 0.55 (95% CI, 0.36-0.83)      Emery & Meeuwisse (2010) (16)  Cluster RCT   Canada  2006-2007  20 weeks Male and female (ages 13-18, tier 1-2, indoor soccer)   INT (380/32) CON (364/28)  Medical attention or time loss  All injury Soccer Injury Prevention Program  - 15-minute warm-up including 5 minutes aerobic and dynamic stretching components and 10 minutes of neuromuscular training (strength, agility, balance).  - 15-minute home-based wobble board balance training program  Overall  INT, 2.08 (95% CI, 1.54-2.74) CON, 3.35 (95% CI, 2.65-4.17)  - IRR all injury, 0.62 (95% CI, 0.39-0.99)  - IRR acute injury, 0.57 (95% CI, 0.35-0.91)       36 Author Study design Country Year Observation period Population (sex, age, level of play, other)  No. of players/no. of teams Injury definition  Targeted injuries Intervention Injury rates (injuries/1,000 player-hours)a   Effect of intervention Soligard et al. (2010) (34)  Prospective cohort study and retrospective survey based on a cluster RCT  Norway  2007  9 months Female (ages 13-17, 15-16 year divisions)  INT (1,055/52)  HC (352/17) IC (351/18)  LC (352/17)  56 coaches   Time loss  All injury FIFA ?11+? - The design, intervention program and main results have been reported by Soligard et al. (2) Teams all injuries HC, 3.1 (95% CI, 2.5-3.8) IC, 3.7 (95% CI, 2.8-4.7) LC, 2.7 (95% CI, 1.6-3.7)   - IRR all injuries, 0.65 (95% CI, 0.46-0.91) - IRR acute injuries, 0.61 (95% CI, 0.42-0.88)  - IRR for all injury HC compared with IC, 0.65 (95% CI, 0.46-0.91)  - IRR for acute injury HC compared with IC, 0.61 (95% CI, 0.42-0.88)      Kiani et al. (2010)(46)  Non-RCT  Sweden  2007  9 months Female (ages 13-19)  INT (777/48) CON (729/49) Traumatic mechanism of injury and medical attention  New acute knee injuries HarmoKnee - Includes warm-up, muscle activation, balance, strength, and core stability - Program performed twice per week during the preseason (12 weeks) and once per week during the regular season INT, 0.04 knee injuries/1,000 player-hours CON, 0.20 knee injuries/1,000 player-hours INT, 0.01 noncontact knee injuries/1,000 player-hours CON, 0.15 noncontact knee injuries/1,000 player-hours  - IRR knee injury, 0.23 (95% CI, 0.04-0.83) - IRR noncontact knee injury, 0.10 (95% CI, 0.00-0.70)     37 Author Study design Country Year Observation period Population (sex, age, level of play, other)  No. of players/no. of teams Injury definition  Targeted injuries Intervention Injury rates (injuries/1,000 player-hours)a  Effect of intervention Wald?n et al. (2012) (47)  Cluster RCT  Sweden  2009  9 months Female (ages 12-17 years, U14 to U18)  INT (2,479/184) CON (2,085/157) Sudden onset and time loss  Acute knee injury Kn?kontroll - 6 exercises: one legged knee squat, pelvic lift, two legged knee squat, bench, lunge and jump landing technique - Exercises were preceded by 5 minutes of low intensity running - Program includes progressions for the exercises and pair exercises to be done with a teammate  INT, 0.33 acute knee injuries/1,000 player-hours CON, 0.36 acute knee injuries/1,000 player-hours  - IRR ACL injury, 0.36 (95% CI, 0.15-0.85)  aUnless stated otherwise; bIncludes overuse injuries and practice injuries; c Only acute injuries were included INT, intervention; CON, control; ACL, anterior cruciate ligament; RR, risk ratio; IRR, incidence rate ratio; AE, athlete exposure; RCT, randomized controlled trial; HC, high compliance tertile; IC, intermediate compliance tertile; LC, low compliance tertile; OR, odds ratio; CI, confidence interval      38 2.4.3 Limitations of previous research Limitations of previous intervention studies include lack of randomization (8,12,33,46), and low adherence to the program.(3) Other limitations include selection bias (8,33,47), low power (8,12,47), inclusion of non-soccer related injuries (46), high drop-out rates (2,8,47), and low injury rates.(46,47) There are also many limitations surrounding the collection of exposure data. Limitations around exposure data collection include failure to record individual exposure hours (3,46), imputation of exposure hours (16), or lack of exposure hours altogether.(12) The majority of the studies that used a cluster design, failed to adjust for clustering in their analyses. Failing to adjust for cluster leads to an underestimation of p-values and a narrower confidence interval, thus increasing the likelihood of finding significant results.(81) Finally, variations in study design, definitions, and methodology have made it difficult to compare results from different studies.   2.5 Rationale, objectives and hypotheses 2.5.1 Rationale As demonstrated in the literature, there is a high incidence of lower-extremity injuries in youth soccer players and strong evidence that neuromuscular training programs can decrease the risk of these injuries. Finch has expanded on van Mechelen?s model of injury prevention by adding two additional stages that focus on developing and understanding the implementation context (stage 5), and determining the effectiveness of the intervention in a real-world context (stage 6).(82)   The 6-stage ?Translating Research into Injury Prevention Practice? (TRIPP) framework has a major focus on implementation and moving efficacy research into direct injury prevention. The majority of injury prevention research in youth soccer corresponds to stage 4 of the TRIPP framework. Although research in stage 4 is important in determining the efficacy of interventions, it does little in providing information on whether or not the results will lead to adoption and success in the real-world.    39   Figure 3 The ?Translating Research into Injury Prevention Practice? framework (82)  Building on recent work by Emery & Meeuwisse (16), this study will evaluate the effectiveness of an injury prevention program when community-driven and delivered.  40 2.5.2 Study 1: The effectiveness of a community-driven injury prevention program in reducing soccer-related injuries in female youth soccer players Objective: To determine if the incidence proportion of time loss injuries of one week or more decreases following the implementation of a neuromuscular training injury prevention program compared to a historical season where no intervention was in place.  2.5.3 Study 2: Risk factors for injury in youth female soccer players  Objective: To examine independent risk factors for all injury and lower-extremity injury in youth female soccer players exposed to an injury prevention program.    41 Chapter  3: Methods  3.1 Purpose To determine whether a community-delivered neuromuscular training program is effective in preventing soccer-related injuries resulting in time loss of one week or more, in a youth female soccer club during one season.   3.2 Research design The study design is a historical cohort study.   3.3 Participants 3.3.1 Sample size The sample size estimation (n = 342 in each study group) is based on an estimated incidence proportion (injuries ? 1 week time loss) in the control group of 15 injuries/100 players and the ability to detect a difference of 50% between the control group and the intervention group (Power [1-?] = 0.8; Alpha [?] = 0.05; Intra-cluster correlation coefficient [?] = 0.02).(16) The sample size must account for the necessity to assess the treatment effect against the between-group variance. Individuals within clusters have the tendency to respond similarly and the natural variability in response among clusters exceeds the variability in response within clusters. This leads to decreased efficiency of cluster randomization relative to individual randomization. To ensure similar power to a study randomizing individuals, the calculated sample size must be adjusted by an ?inflation factor? approximately equal to 1 + (m-1)?, where m is the average cluster size and ? is the value of the intra-class correlation factor. The intra-cluster correlation reflects the within cluster resemblance anticipated. The intra-cluster correlation (? = 0.02) was based on a previous RCT in a similar population.(16) The calculations based on Donner and Klar are found in Appendix A.(83)  3.3.2 Recruitment During the 2008-2009 season, all competitive youth female soccer teams and individuals from the Semiahmoo Soccer Club were invited to participate in the collection of quality assurance data. (N = 23 teams; n = 352 players). The 2008-2009 data was obtained through  42 individual opt-out consent. Player consent to participate in the study was implied if the player and parent did not opt-out of the study. One player chose to opt out of the study and therefore was excluded from the data collection process. A secondary analysis was performed on the quality assurance data collected during the 2008-2009 season.    During the 2010-2011 season, all competitive youth female soccer teams and individuals from the Semiahmoo Soccer Club were invited to participate in the study (N = 20 teams; n = 323 players). Recruitment took place during August and September of 2010, prior to the start of the season. Players were recruited into the study once they returned the completed consent form signed by both the player and parent. All players on participating teams were exposed to the intervention but only those who provided consent were evaluated in this study.   The recruited teams for both seasons played in the top three divisions in various leagues in South Surrey and the Greater Vancouver area. Teams in the U13 to U18 age categories played in Silver, Gold or Metro divisions while the U11 and U12 teams played in the Selects division.  3.3.3 Inclusion and exclusion criteria  3.3.3.1 2008-2009  Criteria for secondary database inclusion: 1. Youth (ages 9-17) female soccer player in the top three divisions of play in the Semiahmoo Soccer Club 2. Participation in the 2008-2009 season  Criteria for secondary database exclusion: 1. Opt-out of the study  43 3.3.3.2 2010-2011  Criteria for study inclusion: 1. Youth (ages 9-17) female soccer player in the top three divisions of play in the Semiahmoo Soccer Club 2. Participation in the 2010-2011 season 3. Completion of signed informed consent by player and parent/guardian  (Appendix B) Criteria for study exclusion: 1. Declined participation 2. Players with an injury resulting in the inability to participate fully at practices or games at the commencement of the season 3. Players with systemic disease including metabolic and/or rheumatologic diseases  3.4 Data collection 3.4.1 Baseline testing Baseline characteristics available for the 2008-2009 cohort include level of play, age group, and age at the beginning of the season. This quality assurance data was obtained from roster information sheets provided by the club.   Baseline testing for the 2010-2011 cohort was done at the beginning of the intervention season (September-November 2010) to examine risk factors for injury in the presence of a community-driven neuromuscular training program. Participants completed a preseason questionnaire (Appendix C) and a Physical Activity Readiness Questionnaire (PAR-Q, Appendix D). The preseason questionnaire asked about age, sex, division of play, soccer position, maturation, physical activity patterns and previous medical history. This questionnaire has been previously validated in an adolescent soccer population.(6) Participants were followed up by telephone to obtain any information that was missing from the preseason questionnaire.   Testing was completed over four days at the Semiahmoo Soccer Club. Twelve teams were tested over two days at the beginning of September 2010, two teams at the beginning of  44 October 2010 and one team at the beginning of November 2010. Five stations were set up indoors and included: anthropometrics; balance; abduction and adduction strength testing; knee extension and flexion strength testing and a vertical drop jump station. Participants began at the height and weight station and were then cycled through the other stations. Once the team had completed the indoor baseline testing, they proceeded outdoors to the artificial turf field for the L?ger 20 metre shuttle run. Two teams did not run the L?ger test because they had participated in a soccer game prior to the baseline testing session.  3.4.1.1 Anthropometrics Height and weight were measured for each participant.   3.4.1.2 Balance testing Unipedal balance was tested using the eyes closed dynamic protocol described by Emery et al.(84) Participants stood barefoot on the Airex Balance Pad (Alcan Airex AG, Sin, Switzerland) with their hands on their hips. They were instructed to close their eyes and elevate one foot off of the pad. Time began when the player lifted their foot off the balance pad and was stopped when the subject lost balance. Loss of balance was defined as: removal of hands from hips; touching down of the non-weight-bearing foot on the pad or floor; movement of the weight-bearing foot from its original position on the pad and movement of the pad itself.(84) The maximum amount of time allowed for the test was 180 seconds. No participants reached the maximum time. Times were recorded to the nearest hundredth of a second using a stopwatch. The maximum time from three trials was used as the variable of interest.   3.4.1.3 Vertical drop jump Vertical drop jump was performed from a 31 cm step (The Step, Reebok Intl Ltd, Canton, MA) and recorded using 2-D video (HDR-SR1 Handycam Camcorder, Sony, Toronto ON). Reflective markers were placed on the bilateral anterior superior iliac spine (ASIS), center of the patella and medial and lateral malleolus. The participant was instructed to hop off the step and  45 perform a two-foot maximal effort jump immediately upon contact with the ground. The jumps were performed with the hands above shoulder height in order to eliminate momentum provided by the arm swing. Each player performed three trials. A physiotherapist classified each trial as low-risk or high-risk (Figure 4) based on dynamic knee valgus angle definitions developed by Ekegren et al.(85) If any of the three trials were classified as high-risk, the player was given a high-risk rating for the vertical drop jump variable. Ekegren et al. have previously reported interrater agreement for this test (Kappa agreement = 0.75-0.85).(85)     Figure 4 Vertical drop jump low-risk (left) and high-risk (right) landing  3.4.1.4 Aerobic fitness The 20 m shuttle run test was used to test aerobic fitness following the protocol described by L?ger et al.(86) The test was performed on an artificial turf field. Consented players ran the test in their teams while two recorders noted the lap at which they were unable to continue. The 20 m shuttle run test has been shown to be both reliable and valid at predicting VO2 max in a youth population with test-retest reliability coefficients of 0.89 for children (ages 6-16 years).(86) In the 20 m shuttle run test, players run back and forth on a marked course in time to a sound signal from an audio recording. Running speed begins at 8.5 km/h and increases by 0.5 km/h per stage.    46 Maximal aerobic shuttle running speed as well as the age of the player were used in the following equation to estimate VO2 max (mL/kg/min) (86):  VO2 max = 31.025 + (3.238?speed) ? (3.248?age) + (0.1536?age?speed)               [Equation 1]  The maximal aerobic running speed was the speed at the last stage completed and age used was the player?s integer age in years (rounded down).  3.4.1.5 Strength testing Strength testing was evaluated by one of two physiotherapists using a handheld force dynamometer. Pillows, straps and a goniometer were used to assist in correct positioning of the subject. Tests were performed for the following movements: hip abduction, hip adduction, knee flexion, and knee extension. For each test the subject was instructed to perform a maximum isometric contraction for 5 seconds, ramping up the force slowly. Each test was performed three times on each side with a short rest in between trials. Peak force values in kilograms were recorded to the nearest tenth of a kilogram using the handheld dynamometer. The average from the three trials was used as the value of interest.   Hip abduction strength was tested using the protocol described by Leetun et al.(87) Hip abduction strength was tested with the subject lying on their side with a pillow between the knees. The hip was positioned in 10 degrees of abduction. A stabilizing strap was placed over the iliac crest and around the table to secure the subject to the table. Another strap was used to hold the force dynamometer in place on the lateral aspect of the thigh. The centre of the dynamometer was positioned 5 cm proximal to the lateral knee joint line.   For hip adduction, the subject was tested while lying on their side with a stabilizing strap over the iliac crest as in the abduction test. The upper leg was flexed to 45 degrees and rested on a pillow. The lower leg was adducted 5  47 degrees and placed on a pillow. The centre of the force dynamometer was positioned 5 cm proximal to the medial knee joint line on the medial thigh and stabilized using a second strap.   Knee extension was tested with the subject seated on the edge of the table with the hip flexed to 90 degrees and the knee flexed to 45 degrees. A stabilizing strap was placed just distal to the ASIS and around the table. A second stabilizing strap was placed around the distal thigh to stabilize the testing leg. The centre of the force dynamometer was placed on the anterior surface of the leg 5 cm proximal to the lateral malleolus and held in place with a third stabilizing strap.   For knee flexion, the subject was tested in the prone position with the knee flexed to 50 degrees. A stabilizing strap was placed around the waist to secure the subject to the table. The centre of the dynamometer was placed on the posterior surface of the leg 5 cm proximal to the lateral malleolus and held in place with a second strap.   Baseline testing was conducted as close to the beginning of the season as possible. All evaluators were given data collection instructions and participated in a training session prior to testing.  3.4.2 Injury surveillance and documentation The injury surveillance system used in this study has been previously validated in a youth soccer population.(6) This injury surveillance system includes the preseason questionnaire (PQ), weekly exposure sheet (WES), injury report form (IRF), therapist assessment form and physician assessment form (Appendix C, Appendix F, Appendix G, Appendix H, Appendix I). Injury information collected during the 2010-2011 season was compared with the secondary database collected by the Safety Director at the Semiahmoo Soccer Club during the 2008-2009 season.    48 3.4.2.1 Operational definitions 1. Injury for the historical control season was previously defined as any soccer injury that resulted in time loss ? 7 days.  Injury for the intervention season was defined as ?any soccer injury that results in the inability to complete a full session, and/or miss a subsequent session, and/or require medical attention.?(40)  2. Injury severity was defined as ?the number of days that have elapsed from the date of injury to the date of the player?s return to full participation in team training and availability for match selection.?(40) a. Slight: 0 days b. Minimal: 1-3 days c. Mild: 4-7 days d. Moderate: 8-28 days e. Severe: > 28 days  3.4.2.2 2008-2009 Injury surveillance The Safety Director (SD) at the Semiahmoo Soccer Club established an independent injury database during the 2008-2009 season that included all injuries resulting in time loss of one week or more. Injuries were documented weekly by the club SD and were followed until return to play. Details surrounding the injury event were collected using a Microsoft Access 2007 survey (Microsoft Corporation, Mississauga, ON) developed by the SD. The survey contained information on location and type of injury, time loss, playing surface, cleat type, time of injury, mechanism, cost and a description of the injury-inciting event (Appendix E). The SD initiated an injury report for those injuries that she was able to capture during weekly follow-up. These reports were completed via phone or email interview. The completed injury reports were sent to the parent contact, of the injured player, to review and edit if necessary. The survey was also sent to all coaches and parents of players at the club. Coaches and parents were asked to retrospectively document any missed injuries that occurred during the  49 previous season. Baseline characteristics (level of play, age group) were obtained through club roster information sheets.   3.4.2.3 2010-2011 Injury surveillance Each team selected a team designate (coach, parent, or manager) to be responsible for completing WESs (Appendix F) and initiating IRFs (Appendix G). The WES contained information on the participation status of all players on the team for each practice and game. Participation status for each player was recorded as full, partial (< 75%) or none. During data entry, partial participation was recorded as half the number of full session hours. Soccer-specific injury information was collected when an injured player presented to the team designate or study physiotherapist. The team designate initiated the IRF while the physiotherapist completed the therapist assessment section of the IRF (Appendix H). The study physiotherapist attended one training session per week to assess injuries and complete the therapist assessment forms. The research coordinator followed up with players by telephone to obtain any missing information pertaining to their injuries. Physician assessment forms (Appendix I) were also distributed but only one was completed during the study. As a result, this data was not reported in this study.   3.5 Outcome measures 3.5.1 Primary outcome Injuries sustained during the 2010-2011 season were documented by the team designate and/or the study physiotherapist using the IRF and therapist assessment form. The team designate completed WESs to document athlete participation hours. The primary outcome was any soccer injury resulting in time loss of one week or more. Soccer injury was defined as any soccer-related injury that resulted in medical attention and/or the removal of the player from the current session and/or subsequent time loss of at least one soccer session (game or practice) as a direct result of that injury.        50 3.5.2 Secondary outcome  All soccer injuries that met the intervention season injury definition regardless of time loss was the secondary outcome of interest. Baseline information was used to determine risk factors for injuries that occurred during the 2010-2011 season.  3.6 Intervention Injury surveillance included data collection with teams in the Semiahmoo Soccer Club (South Surrey, BC) over two seasons: 2008-2009 and 2010-2011. Prior to the 2010-2011 season the Semiahmoo Soccer Club independently decided to implement an injury prevention program to address the issue of lower-extremity injuries in their youth female soccer population. The injury prevention program was initiated, taught and delivered by members of the Semiahmoo Soccer Club community. The SD of the Semiahmoo Soccer Club, was the main team member involved in implementing the injury prevention program at the club. Following observations on youth soccer, footwear, playing surfaces and injuries the SD developed a set of questions to explore the incidence of injury at the Semiahmoo Soccer Club. The SD collected quality assurance data that documented time loss injuries of one week or more during the 2008-2009 season. She noticed a high number of lower-extremity injuries occurring at the club. The SD contacted Dr. Carolyn Emery to learn more about injury prevention in youth soccer.   The SD continued to collect quality assurance data and in November 2009, Dr. Emery presented a lecture on injury prevention in youth female soccer players and the importance of neuromuscular training. The workshop reviewed the evidence associated with risk factors for injury and the effectiveness of the ?Soccer Injury Prevention Program? (SIPP) at reducing injury in youth female soccer players. The presentation was given to coaches, managers and parents affiliated with the club.   In May 2010, Dr. Emery approached the Semiahmoo Soccer Club to find out if there was interest in participating in a formal study assessing the effectiveness of a community-delivered injury prevention program given that the club wanted to deliver the SIPP. The club agreed to participate in the study and the collaboration began in the summer of 2010. The  51 intervention took place during the 2010-2011 season. The 2008-2009 season will serve as the control season.  3.6.1 Personnel and facilities Injury prevention resources were created to distribute to members of the Semiahmoo Soccer Community. One team from the club was featured in a video that demonstrates the various components of the injury prevention program. The coach of each team received a DVD copy (Figure 5) of the injury prevention program. Exercise cards (Appendix J) and brochures (Appendix K) were produced to distribute to the coaches and players.     Figure 5 ?Soccer Injury Prevention Program? DVD  The SD was responsible for distributing the resources to the coaches and teams. This included the DVDs demonstrating the various components of the program as well as ensuring that all the players received their wobble boards and laminated instructions for the home-portion of the program. The SD and a personal trainer were responsible for teaching the players and coaches the team-based portion of the program at the beginning of the season. Following the introduction of the program, the SD and personal trainer frequently attended warm-up sessions to monitor compliance to the program and proper technique while performing the exercises. The warm-up program was trainer-led for the first month with the player designate and/or coach taking over the delivery of the program following this introductory period. Compliance to the warm-up program and quality of drills decreased when the player designate and/or coach took the lead in October 2010. As a result, the  52 program was reassessed and some adjustments were made to the warm-up program based on coach feedback. The program resumed being trainer-led for the 6-weeks prior to the Christmas break. The player designate and/or coach returned to leading the warm-up in January 2011 and continued to lead the program for the remainder of the season.   3.6.2  Injury prevention program The SIPP was implemented at the end of August 2010. The intervention is an active primary prevention program that is composed of both a team and individual component (Appendix J). The SIPP has been proven to be efficacious in a youth soccer population using a randomized controlled trial study design.(16) Emery & Meeuwisse found that use of the SIPP was efficacious in decreasing the risk of all injuries by 38% and acute injuries by 43%.(16)  The team component is a 15-minute warm-up that includes aerobic exercises, dynamic stretching, strength, agility and soccer-specific balance exercises (Table 5). The SIPP was designed to replace the usual team warm-up routine. The team component was to be performed before each training session and game. During the 2010-2011 season, the warm-up was done with two to three other teams on practice days. The only equipment necessary for the team-based component of the injury prevention program was a soccer ball for each player.   The individual component of the intervention is a 15-minute home based wobble-board training program to be performed three times per week (Table 6). Each player received a wobble board (16? diameter, Fitter first, Calgary, Alberta) at the beginning of the season.   Feedback from players and coaches resulted in changes being made to the team component of the program. These changes were made to increase efficiency of the warm-up and to keep the players warm and engaged. The changes to the program were made as a result of collaborations between the SD, personal trainer and Dr. Emery. Changes were made to the dynamic stretching and strength exercises of the program. Leg swings were followed by ?open and close the gate? dynamic exercises. Abdominal and hamstring strengthening exercises (exercises 12 & 14) were removed from the team warm-up and incorporated into  53 practice. A series of sprints were added onto the end of the warm-up to ensure the players were warm as they moved into the tactical portion of their practice. Sprints were done at roughly 70%, 80% and 100% effort to the 30-metre cone line with a 30-metre jog back between each sprint. Plank exercises (exercises 10 & 11) were incorporated into the progression sprints at the end of the warm-up. The soccer-specific balance exercises (exercises 19 & 20) were done as a team once the team had moved into their individual practice environment and began ball work. It is important to note that the soccer club and not the research team were responsible for initiating these changes to the prescribed injury prevention program. These changes were deemed appropriate and necessary in order to keep the teams participating in the injury prevention program.  54 Table 5 Exercises and repetitions of the injury prevention program used as a structured warm-up program. See Appendix J for pictures. Exercise Focus Repetitions Aerobic component   Forward run  Relaxed running technique, using arms, knee over ankle, progress speed on 2nd repetition.  30 metres x 2    Backward run  Relaxed running technique, bending the knees and hips, landing on toes, knee over ankle, progress speed on 2nd repetition.  30 metres x 2    High knees  Driving each knee high in front of the hip and landing lightly on toes with good alignment of the knee over ankle not allowing the knee to fall inwards. 30 metres    Heel kicks  Knees bending so the heel touches the buttocks, landing lightly on the toes with good alignment of the knee over the ankle. 30 metres    Power skip  Using a burst of power in the take-off leg while drawing the opposite arm forward and up. Landings should be light on the toes absorbing with the knee bent and maintaining good alignment of the knee over the ankle.  30 metres x 2    Zig zag shuffle  Pushing from the back leg with good alignment of the knee over the ankle. Do not let knee fall inward particularly on the change in direction.  30 metres x 2    Dynamic stretching   Leg swings front & back  Maintaining good control of the pelvis without excessively arching or flexing the back. Movement should be slow and through maximal comfortable range of motion.  2 x 10 reps    Leg swings side-to-side  Alternating crossovers in front and behind other leg. Slowly through maximal comfortable range.  2 x 10 reps    Calf stretch walk  Knees straight as possible, heels on ground.  2 x 10 reps    Strength   Plank- elbows  Engaging abdominal stabilizing muscles. Elbows under shoulders, body should form a straight line. Tightening abdominal muscles by pulling the abdominals inward. 1 minute; 15-60 second holds (progressive)     55 Exercise Focus Repetitions Side plank- elbow  Elbow under shoulder, body should form a straight line. Tightening of the abdominal muscles.  1 minute; 15-60 second holds (progressive)    Abdominal exercise  Hips and knees at 90 degrees to start, alternate legs slowly. Engage abdominal muscles. Lower back on ground. 1 minute; 15-60 seconds (progressive)    Walking lunge  Knee no greater than 90 degrees, slow controlled lowering into lunge, knee alignment over ankle, do not let knee fall inward. Maintain abdominal stabilization. 30 metres    Eccentric hamstring  Lower slowly, keeping body in a straight line while engaging the abdominal muscles.  5-10 reps    Calf raises  Lower slowly, feather-touch heels to ground. Double leg progressing to single leg.  3 x 10 reps    Agility   Two-legged jumps  (forward & back, side-to-side) Soft landing on toes, absorbing and lowering to heels. Take off and landing with hips and knees bent. Maintaining knee over ankle alignment, avoiding the knee falling inward. Engaging abdominal muscles.  10-20 reps of each type    Single-leg jump  (forward & back, side-to-side) Soft landing on toes, absorbing through bent knee and hip with proper alignment. Progressed by achieving greater height and distance.  10-20 reps of each type    Skate jumps  Take-off with one foot, knee aligned over the ankle followed by landing on the other foot across the line. Hip directly over ankle on landing, avoid reaching out. 20 reps    Balance   Single leg balance toss  Maintain a bent knee and knee aligned over ankle. Avoid knee falling inward.  30 seconds each leg    Single leg passing  Maintain bent knee over ankle. Balance continuously on the non-kicking leg. 30 seconds each leg    Single leg jousting  Intermittently push partner in different directions, maintain balance. Knee over ankle. 60 seconds each leg    Single leg juggling  Try not to switch legs. 60 seconds each leg  56 Table 6 Exercises and repetitions for the home-based wobble board component of the injury prevention program. See Appendix J for pictures. Exercise Focus Repetitions Weeks 1 & 2   Dynamic balance forward & back- double legged  Feet parallel, knees slightly bent. Feather touch edge to surface, knees over ankles. Repeat slowly and continuously. 5 reps, 30 seconds    Dynamic balance side to side- double legged  Feather touch edge to surface, knees over ankles. Repeat slowly and continuously. 5 reps, 30 seconds    Dynamic balance circles clockwise/ counterclockwise- double legged  Feather touch edge to surface, knees over ankles. Repeat slowly and continuously.  5 reps each direction, 30 seconds    Static balance  Stand with one foot centered on the wobble board. Knee slightly bent. Maintain board parallel to surface. 10 reps, 10 seconds    Static balance eyes closed  Close eyes for last 5 seconds. Maintain board parallel to surface. 10 reps, 10 seconds    Weeks 3 & 4      Dynamic balance forward & back- single-leg  Feather touch edge to surface, knee over ankle. 5 reps, 15 seconds    Dynamic balance side to side- single-leg  Feather touch edge to surface, knee over ankle. 5 reps, 15 seconds    Dynamic balance clockwise/ counterclockwise- single-leg  Feather touch edge to surface, knee over ankle. 5 reps, 15 seconds    Static balance  Stand with one foot centered on the wobble board. Knee slightly bent. Maintain board parallel to surface. 10 reps, 20 seconds    Static balance eyes closed  Close eyes for last 5 seconds. Maintain board parallel to surface. 10 reps, 20 seconds    Advanced Wobble Board Drills   Jousting Intermittently push partner in different directions, maintain balance. Knee over ankle.     Wall or partner ball toss Maintain a bent knee and knee aligned over ankle. Avoid knee falling inward.    57 3.7 Statistical analyses 3.7.1 Predictive and descriptive statistics Stata 12.0 (Statacorp, College Station, TX) was used to perform all statistical analyses. All tests were two-sided using a p-value of 0.05. Study Trax (ScienceTrax, Macon, GA) was used for data entry, storage and organization.   Detailed statistical analyses plans are outlined within each study chapter.    58 Chapter  4: The effectiveness of a community-driven injury prevention program in reducing soccer-related injuries in female youth soccer players  4.1 Introduction Injury incidence rates reported in youth female soccer are 1 to 5 injuries per 1,000 practice-hours and 8 to 22 injuries per 1,000 game-hours.(2-7) Prospective cohort studies report that lower-extremity (LE) injuries account for 67% to 100% of all injuries sustained in youth soccer players.(2-6,8-16) In youth female soccer players, the most commonly injured body parts are the knee (14-32%), ankle (23-35%) and upper leg (8-21%).(2-5,12) This has led to injury prevention efforts focused on ameliorating risk factors that are associated with lower-extremity injuries in youth soccer.   Six studies have assessed the effectiveness of multifaceted neuromuscular training injury prevention programs in youth female soccer players.(2,3,12,16,33,34) Of those, five reduced injuries (2,12,16,33,34) while one found no difference in injury rates between the intervention and control groups.(3) These injury prevention programs reduced the risk of overall injury (2,12,16), anterior cruciate ligament (ACL) injury (33), severe injury (2), acute-onset injury (16), and overuse injury (2) in youth female soccer players.   While there is clear support for the use of neuromuscular training to prevent injury in youth female soccer players, all of the studies that have provided this evidence correspond to stage 4 of the ?Translating Research into Injury Prevention Practice? (TRIPP) framework developed by Finch.(82) This stage of the TRIPP framework focuses on evaluation of injury prevention programs under ideal conditions (e.g., randomized controlled trials [RCT]), which contributes to the evidence base around the efficacy of neuromuscular training in preventing injury in youth female soccer players. However, there is a paucity of effectiveness and implementation research evaluating how these interventions operate and succeed in the real-world. Effectiveness trials and implementation research correspond to TRIPP stages 5 and 6, the final two stages of the framework.(82) The community setting is the natural mode of delivering injury prevention programs and the focus of injury prevention research in youth  59 soccer now needs to shift towards knowledge translation and how injury prevention programs can be best applied in a real-world setting.   Emery and Meeuwisse provide further RCT support for neuromuscular training for the prevention of injuries in youth soccer.(16) The ?Soccer Injury Prevention Program? (e.g., neuromuscular and balance training program) implemented by Emery & Meeuwisse decreased the risk of all injuries by 38% and acute-onset injuries by 43%.(16) This is consistent with a RCT study in Norway where Soligard et al. also demonstrated a protective effect of a neuromuscular training program (?11+?) at reducing all injuries (IRR, 0.68; 95% CI, 0.48-0.98), overuse injuries (IRR, 0.47; 95%CI, 0.26-0.85), and severe injuries (IRR, 0.55; 95% CI, 0.36-0.83) in youth female soccer players.(2)   The efficacy of the ?Soccer Injury Prevention Program? (SIPP) is understood but the effectiveness of this specific program when driven and delivered by a soccer community is unknown. This study provided a unique opportunity to evaluate a community-driven injury prevention program and to compare the results with the RCT efficacy results reported by Emery & Meeuwisse.(16) Therefore, the objective of this study was to examine the effectiveness of a community-driven injury prevention program in reducing time loss injury in youth female soccer players. We hypothesized that the effectiveness of a community-driven injury prevention program would be similar in magnitude to the efficacy studies reported in the literature.   4.2 Methods The recruitment process and methods are outlined in chapter 3. All players from both cohorts (2008-2009 & 2010-2011) were included for analysis in this study. The injuries of interest were those with time loss ? 7 days. The time loss definition used was a result of the definition used to collect injuries in the secondary database during the 2008-2009 season.   60 4.2.1 Statistical analysis 4.2.1.1 Descriptive epidemiology Descriptive statistics were used to compare baseline characteristics between cohorts as well as injured players compared to non-injured players within each cohort. Categorical variables are presented as frequencies and percentages. Continuous variables are presented using means and 95% confidence intervals. Additional baseline characteristics are provided for the 2010-2011 cohort.   Descriptive statistics were used to describe characteristics of the injuries. Injury location, type, severity, mechanism and time of injury were described using frequencies and percentages. Definitions for severity were obtained from the Consensus Statement on Injury Definitions and Data Collection Procedures in Studies of Soccer Injuries.(40)   4.2.1.2 Risk ratio Risk ratios were estimated based on incidence proportions in each cohort to assess the effectiveness of the SIPP on injury risk through comparison of the control season and the intervention season. The incidence proportion is the number of injuries that occurred within the study period divided by the population size and is presented as injuries per 100 players. Incidence proportion was estimated for all injury, acute-onset injury and lower-extremity injury and stratified by cohort and age group. Only injuries with time loss ? 7 days were included in the effectiveness analysis as a result of the injury definition used during the control season.    Incidence proportions and unadjusted risk ratio (RR, with 95% confidence intervals) were estimated for all injury, acute-onset injury and lower-extremity injury using Poisson Regression Analysis adjusted for clustering by team.(81) RRs were estimated for each cohort independently and used to compare injury risk in the intervention season to that in the control season. Effect modification was examined using a stratified analysis to compare the incidence proportion across the four age groups. The adjusted RR was estimated using Poisson Regression Analysis adjusting for cluster and other covariates (e.g., age group, level  61 of play). Adjusted RRs were used to assess secondary risk factors for injury, independent of study cohort. The secondary risk factors of interest were age group and level of play.   An exploratory analysis was performed to assess the effect of the intervention on knee injury, ankle injury, knee sprain injury and ankle sprain injury. This data is exploratory since the study was not powered to detect reductions in specific injury types. All tests were two-sided using a p-value of 0.05.   4.3 Results 4.3.1 Baseline characteristics There were 351 players in the control cohort (2008-2009) and 187 players in the intervention cohort (2010-2011). Figure 6 presents recruitment of players into the study. The mean age of the 2008-2009 cohort was 14.0 years (95% CI, 13.8-14.2) while the mean age of the 2010-2011 cohort was 13.1 years (95% CI, 12.9-13.4). Ninety-six players played during both the 2008-2009 and 2010-2011 seasons and therefore appear in both the control cohort and the intervention cohort. Accounting for the overlap in players, the independent sample size for the study was 442 players.              Target population 2008-2009 N = 23 teams (n = 352 players)  Target population 2010-2011 N = 20 teams (n = 323 players)  Declined to participate n = 1 player Declined to participate n = 136 players Recruited players 2008-2009 N = 23 teams (n = 351 players)  Recruited players 2010-2011 N = 15 teams (n = 187 players)  Total number of players included in analyses N = 38 teams (n = 538 players)  Figure 6 Flow chart describing recruitment of players into the study.  62 Table 7 presents the comparison of baseline characteristics between players from the two cohorts. The intervention season had a higher proportion of players in the Tier 2 group and a smaller proportion of players in the Tier 3 group compared to the control season. The proportion of players in the Tier 1 group was similar between the cohorts. In the intervention season, there were more players in the U14 and U16 age groups and fewer players in the U18 age group compared to the control season.  Table 7 Baseline characteristics by cohort   No. (%)a  Cohort Characteristic 2008-2009 n = 351 2010-2011 n = 187 Age, mean (95% CI), years 14.0 (13.8-14.2) 13.1 (12.9-13.4) Level of play   Tier 1 103 (29.3) 61 (32.6) Tier 2 156 (44.4) 102 (54.5) Tier 3 92 (26.2) 24 (12.8) Age group   U12 60 (17.1) 46 (24.6) U14 97 (27.6) 65 (34.8) U16 97 (27.6) 68 (36.4) U18 97 (27.6) 8 (4.3) aUnless stated otherwise; CI, confidence interval  63 Table 8 presents the comparison of baseline characteristics between players from the two cohorts by injury status. In the control season, injured players tended to be older and in the highest level of play at baseline compared to uninjured players. Likewise, during the intervention season injured players tended to be in the highest level of play and older age group at baseline compared to those players that were not injured.   Table 8 Baseline characteristics by cohort and injury status  No. (%)a  Injured players Uninjured players Characteristic 2008-2009 n = 50 2010-2011 n = 19 2008-2009 n = 301 2010-2011 n = 168 Age, mean (95% CI), years 15.0 (14.4-15.6) 13.2 (12.3-14.1) 13.8 (13.6-14.1) 13.1 (12.9-13.4) Level of play     Tier 1 24 (48.0) 8 (42.1) 79 (26.3) 53 (31.5) Tier 2 21 (42.0) 9 (47.4) 135 (44.9) 93 (55.4) Tier 3 5 (10.0) 2 (10.5) 87 (28.9) 22 (13.1) Age group     U12 4 (8.0) 4 (21.1) 56 (18.6) 42 (25.0) U14 8 (16.0) 6 (31.6) 89 (29.6) 59 (35.1) U16 15 (30.0) 8 (42.1) 82 (27.2) 60 (35.7) U18 23 (46.0) 1 (5.3) 74 (24.6) 7 (4.2) aUnless stated otherwise; CI, confidence interval  64 Table 9 provides additional baseline characteristics that were collected for the 2010-2011 cohort. This table also compares additional baseline characteristics between those players that were injured during the 2010-2011 season and those that were not injured during the 2010-2011 season.  Table 9 Additional baseline characteristics for the 2010-2011 cohort Characteristic All players n = 187 Injured players n = 19 Uninjured players n = 168 Mean (95% CI)     Height, cm 157.3 (155.9-158.7) 158.4 (152.2-164.5) 157.2 (155.8-158.6) Missing data (%) 10 (5.3) 2 (10.5) 8 (4.8) Weight, kg 48.8 (47.3-50.4) 50.4 (44.4-56-5) 48.6 (47.0-50.3) Missing data (%) 10 (5.3) 2 (10.5) 8 (4.8) BMI, kg/m2 19.5 (19.1-19.9) 19.9 (18.4-21.3) 19.5 (19.1-19.9) Missing data (%) 10 (5.3) 2 (10.5) 8 (4.8) Sports participation, hours/week 6.9 (6.4-7.4) 6.2 (4.7-7.8) 7.0 (6.5-7.6) Missing data (%) 17 (9.1) 2 (10.5) 15 (8.9) Aerobic fitness, mL/kg/min 47.0 (46.3-47.7) 43.8 (40.6-47.1) 47.2 (46.5-47.9) Missing data (%) 74 (39.6) 11 (57.9) 63 (37.5) Max balance right, s 5.4 (4.9-5.9) 5.7 (3.6-7.8) 5.4 (4.9-5.9) Missing data (%) 10 (5.3) 2 (10.5) 8 (4.8) Max balance left, s 5.1 (4.6-5.5) 4.8 (3.9-5.7) 5.1 (4.6-5.6) Missing data (%) 11 (5.9) 2 (10.5) 9 (5.4) No. (%)    Position    Forward 43 (23.0) 2 (10.5) 41 (24.4) Midfield 61 (32.6) 6 (31.6) 55 (32.7) Defense 52 (27.8) 8 (42.1) 44 (26.2) Goal keeper 11 (5.9) 0 11 (6.5) Missing data (%) 20 (10.7) 3 (15.8) 17 (10.1) Previous injury 1 year     Yes 59 (31.6) 9 (47.4) 50 (26.7) No 111 (59.4) 8 (42.1) 103 (61.3) Missing data (%) 17 (9.1) 2 (10.5) 15 (8.9) VDJ, high-risk rating if high-risk on any trial    High-risk 64 (34.2) 3 (15.8) 61 (36.3) Low-risk 76 (40.6) 10 (52.6) 66 (39.3) Missing data (%) 47 (25.1) 6 (31.6) 41 (24.4) VDJ, high-risk rating if high-risk on all 3 trials    High-risk 30 (16.0) 2 (10.5) 28 (16.7) Low-risk 110 (58.8) 11 (57.9) 99 (58.9) Missing data (%) 47 (25.1) 6 (31.6) 41 (24.4) CI, confidence interval; BMI, body mass index; VDJ, vertical drop jump   65 4.3.2 Descriptive epidemiology 4.3.2.1 Incidence proportion During the 2008-2009 season 50 players sustained 56 injuries of one week or more time loss. Forty-four players sustained a single injury over the season while 6 players sustained 2 injuries. No player sustained more than 2 injuries resulting in time loss of one week or more. The overall incidence proportion for the control season was 16.0 injuries per 100 players (95% CI, 9.9-25.7). Of the 56 injuries, 48 were acute-onset injuries, 4 were overuse and the causes of 4 were unknown. The incidence proportion for acute-onset injuries was 13.7 injuries per 100 players (95% CI, 8.5-22.1). 78.6% of all injuries occurred to the LE resulting in an incidence proportion of 12.5 LE injuries per 100 players (95% CI, 7.3-21.5).  During the 2010-2011 season 19 players sustained 19 injuries of one week or more time loss. No player sustained more than one injury resulting in time loss of one week or more. The overall incidence proportion for the intervention season was 10.2 injuries per 100 players (95% CI, 6.9-15.0). Of the 19 injuries, 15 were acute-onset, 3 were overuse and the cause of 1 was unknown. The incidence proportion for acute-onset injuries was 8.0 injuries per 100 players (95% CI, 4.8-13.4). 78.9% of injuries occurred to the LE resulting in an incidence proportion of 8.0 LE injuries per 100 players (95% CI, 5.2-12.3).  Table 10 provides a summary of the incidence proportion for all injury, acute-onset injury and LE injury for each cohort by age group. In our risk factor analyses, the U12 age group was used as the reference group as it was hypothesized that older age would be a risk factor for injury. The incidence proportion point estimates suggest that incidence proportion increases in older age groups, independent of cohort.    66 Table 10 Incidence proportion by cohort and age group  2008-2009 2010-2011  No. of injuries Incidence proportion (95% CI), injuries/100 players No. of injuries Incidence proportion (95% CI), injuries/100 players All injury     U12 4 6.7 (1.6-27.8) 4 8.7 (3.9-19.4) U14 9 9.3 (4.8-18.1) 6 9.2 (5.6-15.2) U16 18 18.6 (6.2-55.8) 8 11.8 (5.2-26.6) U18 25 25.8 (14.0-47.4) 1 12.5 (n/a) Overall 56 16.0 (9.9-25.7) 19 10.2 (6.9-15.0) Acute injury     U12 4 6.7 (1.6-27.8) 3 6.5 (3.5-12.3) U14 8 8.3 (3.6-18.8) 4 6.2 (2.4-15.5) U16 14 14.4 (4.5-45.9) 7 10.3 (4.0-26.6) U18 22 22.7 (12.4-41.4) 1 12.5 (n/a) Overall 48 13.7 (8.5-22.1) 15 8.0 (4.8-13.4) LE injury     U12 3 5.0 (1.4-18.5) 4 8.7 (3.9-19.4) U14 5 5.2 (3.4-7.8) 4 6.2 (2.9-13.0) U16 16 16.5 (4.9-55.8) 6 8.8 (3.7-21.0) U18 20 20.6 (11.5-36.9) 1 12.5 (n/a) Overall 44 12.5 (7.3-21.5) 15 8.0 (5.2-12.3) CI, confidence interval; LE, lower-extremity; n/a, not available  4.3.2.2 Injury location Table 11 provides a breakdown of all injuries resulting in time loss ? 7 days by location of injury. Ankles and knees were the most common body parts injured for both cohorts. During the 2008-2009 season the most common locations of injury were the ankle (30.4%), knee (16.1%) and upper leg (10.7%). Likewise, the most common locations of injury for the 2010-2011 season were the knee (31.6%), ankle (15.8%) and upper leg (15.8%).  4.3.2.3 Injury type Table 11 provides a breakdown of all injuries resulting in time loss ? 7 days by type of injury. During the 2008-2009 season, the most common types of injury were joint/ligament sprain (41.1%), muscle strain (28.6%) and fracture (7.1%). During the 2010-2011 season, the most common types of injury were joint/ligament sprain (30.0%), muscle strain (20.0%) and contusion (15.0%). Concussion accounted for 5.4% and 5.0% of all injuries that occurred during the 2008-2009 and 2010-2011 seasons respectively.   67 Table 11 Injury location and injury type by cohort  No. (%)  2008-2009 2010-2011 Location   Ankle 17 (30.4) 3 (15.8) Knee 9 (16.1) 6 (31.6) Upper leg 6 (10.7) 3 (15.8) Hip 4 (7.1) - Groin 3 (5.4) - Wrist 3(5.4) - Head 3 (5.4) 1 (5.3) Foot/toes 3 (5.4) 2 (10.5) Lower leg 2 (3.6) 1 (5.3) Back 2 (3.6) 1 (5.3) Finger 1 (1.8) - Ribs - 1 (5.3) Upper arm - 1 (5.3) Unknown 3 (5.4) -    Injury typea   Joint/ligament sprain 23 (41.1) 6 (30.0) Muscle strain 16 (28.6) 4 (20.0) Fracture 4 (7.1) 1 (5.0) Contusion 3 (5.4) 3 (15.0) Concussion 3 (5.4) 1 (5.0) Dislocation - 1 (5.0) Other 3 (5.4) 4 (20.0) Unknown 4 (7.1) - aone injury for the 2010-2011 cohort was categorized as both a knee dislocation and a knee ligament sprain  4.3.2.4 Injury severity Table 12 presents injury severity by cohort. We were only able to include injuries with time loss ? 7 days due to the injury definition used during the control season. Injuries were classified as (40):  ! Mild: 7 days  ! Moderate: 8-28 days ! Severe: > 28 days   68 During the 2008-2009 season, 15 (26.8%) injuries were classified as mild, 21 (37.5%) as moderate, 19 (33.9%) as severe and the severity of 1 (1.8%) injury was unknown. In comparison, during the 2010-2011 season, 4 (21.1%) injuries were classified as mild, 9 (47.4%) as moderate and 6 (31.6%) as severe.   Table 12 Injury severity by cohort  No. (%)  2008-2009 2010-2011 Mild: 7 days 15 (26.8) 4 (21.1) Moderate: 8-28 days 21 (37.5) 9 (47.4) Severe: > 28 days 19 (33.9) 6 (31.6) Unknown 1 (1.8) -  4.3.2.5 Injury mechanism and session type Table 13 provides information on the injury mechanism and session type when the injury took place. During the control season 78.6% of all injuries involved contact with another player while 63.2% of all injuries in the intervention season involved contact with another player. 10.7% of all injuries during the control season were noncontact while 31.6% of all injuries during the intervention season were noncontact. The mechanism of injury was unknown for 10.7% and 5.3% of all injuries for the 2008-2009 and 2010-2011 seasons respectively.   Throughout the 2008-2009 season, 6 (10.7%) injuries took place during practice, 27 (48.2%) during games, and the session type was unknown for 23 (41.1%) injuries. During the 2010-2011 season, 5 (26.3%) injuries took place during practice, 13 (68.4%) during games, and the session type was unknown for 1 (5.3%) injury.    69 Table 13 Injury mechanism and session type by cohort  No. (%)  2008-2009 2010-2011 Mechanism   Contacta 44 (78.6) 12 (63.2) Noncontact 6 (10.7) 6 (31.6) Unknown 6 (10.7) 1 (5.3) Session type   Game 27 (48.2) 13 (68.4) Practice 6 (10.7) 5 (26.3) Unknown 23 (41.1) 1 (5.3) aContact between players   4.3.3 The effectiveness of the SIPP Incidence proportions and risk ratios were estimated using Poisson Regression Analysis adjusted for cluster by team.    4.3.3.1 Incidence proportion. Is the intervention effective at reducing injury? Table 14 provides the unadjusted RR for all injury, acute-onset injury and LE injury. Point estimates suggest that exposure to the injury prevention program reduced the risk of all injury, acute-onset injury and LE injury; however, these associations did not reach significance.  Table 14 Effect of the SIPP on all injury, acute-onset injury and lower-extremity injury  Outcome Injury type No. of injuries Incidence proportion (95% CI), injuries/100 players Risk ratio (95% CI) All injury    2008-2009 56 16.0 (9.9-25.7) 1 2010-2011 19 10.2 (6.9-15.0) 0.64 (0.35-1.17) Acute-onset injury    2008-2009 48 13.7 (8.5-22.1) 1 2010-2011 15 8.0 (4.8-13.4) 0.59 (0.29-1.17) LE injury    2008-2009 44 12.5 (7.3-21.5) 1 2010-2011 15 8.0 (5.2-12.3) 0.64 (0.32-1.26) 2008-2009, n = 351 players; 2010-2011, n = 187 players; SIPP, Soccer Injury Prevention Program; CI, confidence interval; LE, lower-extremity; Adjusted for clustering by team, unadjusted for covariates     70 4.3.3.2 Sensitivity analysis A sensitivity analysis was performed to assess whether the exclusion of the participants that appeared in both cohorts gave similar results as to when they were included in the analyses. Table 15 provides the unadjusted RR (with the players removed) for all injury, acute-onset injury and LE injury, using Poisson Regression Analysis adjusted for cluster. There was little difference in the incidence proportion or RR point estimates for all injury, acute-onset injury and LE injury when the 96 players that appeared in both cohorts were excluded from analyses compared to when they were included in the analyses. Results indicate that our findings are robust to the overlap in players between the two cohorts. Therefore, all of the players were included as independent participants in the analyses discussed in this chapter.  Table 15 Effect of the SIPP on all injury, acute-onset injury and lower-extremity injury with players that appear in both cohorts removed  Outcome Injury type No. of injuries Incidence proportion (95% CI), injuries/100 players  Risk ratio (95% CI) All injury    2008-2009 43 16.9 (10.1-28.2) 1 2010-2011 10 11.0 (6.1-19.8) 0.65 (0.30-1.41) Acute-onset injury    2008-2009 35 13.7 (8.0-23.5) 1 2010-2011 8 8.8 (4.4-17.5) 0.64 (0.27-1.51) LE injury    2008-2009 35 13.7 (8.0-23.4) 1 2010-2011 8 8.8 (4.7-16.6) 0.64 (0.28-1.45) 2008-2009, n = 255 players; 2010-2011, n = 91 players; SIPP, Soccer Injury Prevention Program; CI, confidence interval; LE, lower-extremity; Adjusted for clustering by team, unadjusted for covariates  4.3.3.3 Effect modification by age group A sub-group analysis was performed to examine effect modification of the intervention across age groups. Table 16 provides a summary of the risk ratio for all injury, acute-onset injury and LE injury by age group. The risk ratio point estimates suggest that the effect of the intervention on the risk of injury is different across age groups and that age may be an effect modifier in this study. Risk ratio point estimates suggest that the injury prevention program has the greatest effect on the reduction of all injury, acute-onset injury and LE injury in  71 players in the older age groups. This study was underpowered to draw conclusions on the interaction between the intervention and age group.   Table 16 Sub-group analysis of Risk Ratio by age group  Risk Ratio (95% CI)  All injury Acute-onset injury LE injury Cohort    Intervention U12 1.30 (0.28-6.11) 0.98 (0.22-4.27) 1.74 (0.41-7.36) Control U12 1 1 1 Intervention U14 0.99 (0.45-2.20) 0.75 (0.23-2.41) 1.19 (0.53-2.67) Control U14 1 1 1 Intervention U16 0.63 (0.17-2.33) 0.71 (0.17-2.96) 0.53 (0.13-2.22) Control U16 1 1 1 Intervention U18 0.49 (0.27-0.88)* 0.55 (0.30-0.999)* 0.61 (0.34-1.08) Control U18 1 1 1 CI, confidence interval; LE, lower-extremity injury; *statistically significant at p < 0.05  4.3.3.4 Incidence proportion. Is the intervention effective at reducing injury after controlling for age group and level of play? The primary risk factor of interest was exposure to the injury prevention program. Secondary risk factors that were examined include age group and level of play. Table 17 summarizes the adjusted risk ratio for injury risk factors, using the adjusted Poisson regression models for all injury, acute-onset injury and LE injury. Point estimates suggest that exposure to the injury prevention program is protective against all injury, acute-onset injury and LE injury; however, this association was not statistically significant.   Older age group and higher level of play were hypothesized to be potential risk factors for each of the three main injury outcomes. Independent of cohort, point estimates suggest a greater risk of injury in the older players when compared to the younger players. There was a greater risk of all injury in the U18 group compared with the U12 group (RR, 3.52; 95% CI, 1.72-7.20) after adjusting for exposure to the intervention and level of play. A similar relationship was seen for acute-onset injury and LE injury. There was no significant difference in injury risk seen between the U14 or U16 players compared to the U12 players. Similarly, the overall injury risk of the Tier 1 players was nearly four times greater compared to the Tier 3 players (RR, 3.79; 95% CI, 1.33-10.77) after adjusting for exposure to the  72 intervention and age group. A similar relationship was seen for acute-onset injury and LE injury. Additionally, there was a greater risk of LE injury in Tier 2 players compared to Tier 3 players (RR, 3.16; 95% CI, 1.03-9.70). There was no significant difference in injury risk seen between the Tier 2 players and the Tier 3 players for all injury or acute-onset injury.   Table 17 Risk factor analyses for all injury, acute-onset injury and lower-extremity injury  Risk Ratio (95% CI) Risk factor All injury Acute-onset injury LE injury Cohort    Control 1 1 1 Intervention 0.73 (0.37-1.45) 0.69 (0.33-1.44) 0.74 (0.34-1.64) Age group    U12 1 1 1 U14 1.49 (0.65-3.40) 1.39 (0.56-3.44) 0.99 (0.45-2.17) U16 2.17 (0.84-5.61) 2.00 (0.75-5.34) 2.08 (0.76-5.70) U18 3.52 (1.72-7.20)* 3.50 (1.66-7.38)* 3.27 (1.66-6.46)* Level of play    Tier 3 1 1 1 Tier 2 2.73 (0.90-8.25) 2.59 (0.72-9.29) 3.16 (1.03-9.70)* Tier 1 3.79 (1.33-10.77)* 3.92 (1.20-12.88)* 3.81 (1.35-10.76)* CI, confidence interval; LE, lower-extremity; *statistically significant at p < 0.05  4.3.3.5 Exploratory analysis An exploratory analysis was performed to assess the effect of the intervention on the risk of knee injury, ankle injury, knee sprain injury, and ankle sprain injury. Table 18 provides incidence proportion and risk ratio values for the exploratory outcome variables. Risk ratio point estimates suggest a reduction in ankle injury, knee sprain injury and ankle sprain injury in the intervention cohort compared to the historical cohort; however, these associations did not reach significance. The risk ratio point estimate for knee injury suggests that the intervention cohort was at greater risk of knee injury compared to the historical cohort.    73 Table 18 Exploratory analysis of specific injuries   Outcome Injury type No. of injuries Incidence proportion (95% CI), injuries/100 players  Risk ratio (95% CI) Knee injury    2008-2009 9 2.6 (1.4-4.7) 1 2010-2011 6 3.2 (1.5-7.0) 1.25 (0.47-3.30) Knee sprain    2008-2009 9 2.6 (1.4-4.7) 1 2010-2011 2 1.1 (0.3-4.0) 0.42 (0.10-1.71) Ankle injury    2008-2009 17 4.8 (2.8-8.4) 1 2010-2011 3 1.6 (0.6-4.6) 0.33 (0.10-1.07) Ankle sprain    2008-2009 14 4.0 (2.2-7.4) 1 2010-2011 3 1.6 (0.6-4.6) 0.40 (0.12-1.35) 2008-2009, n = 351 players; 2010-2011, n = 187 players; CI, confidence interval  4.4 Discussion  The objective of this study was to assess the effectiveness of a community-driven injury prevention program at reducing injuries with time loss ? 7 days in youth female soccer players. Point estimates suggest that this training program was effective in reducing all injury, acute-onset injury, and LE injury although these findings did not reach significance. Although not statistically significant we believe that these results are clinically relevant.   4.4.1 Descriptive epidemiology 4.4.1.1 Incidence proportion The overall observed incidence proportion (time loss injuries ? 7 days) in the control group was 16.0 injuries (95% CI, 9.9-25.7) per 100 players. The observed incidence proportion is in agreement with previous studies that have found overall incidence proportions of 4.9 to 34.0 injuries per 100 players when including injuries of similar time loss.(2-6,16) Previous studies have used varying definitions for severity of injury and subsequent time loss, which may explain the wide range of values seen in the literature. Studies by Steffen et al. (3), Emery & Meeuwisse (16), Soligard et al. (2), and Emery et al (6), used time loss ? 8 days to differentiate minor injuries from moderate injuries where studies by Le Gall et al. (5), and  74 Soderman et al. (4) used time loss ? 7 days as the cut-off between a minor and moderate injury. We argue that a one-day difference in time loss is small and that our incidence proportion outcomes support the results seen in the literature. The overall observed incidence proportion in the intervention group was 10.2 injuries (95% CI, 6.9-15.0) per 100 players. This is similar to previous injury prevention intervention studies that have found between 3.7 to 10.4 injuries per 100 players in the intervention groups.(2,3,16) Differences between our results and previous studies may be a result of the injury definition and data collection procedures.  Lower-extremity (LE) injuries accounted for the majority of injuries in both cohorts. The overall observed LE incidence proportion in the control cohort was 12.5 injuries (95% CI, 7.3-21.5) per 100 players compared to 8.0 injuries (95% CI, 5.2-12.3) per 100 players in the intervention cohort. LE incidence proportions have been reported to be 16.5 to 18.3 and 11.1 to 16.9 LE injuries per 100 players per season for control cohorts and intervention cohorts respectively.(2,3,16) Disparity between our results and the literature are likely a result of the difference in injury definitions. The LE incidence proportions available in the literature include all injuries, regardless of time loss, whereas our study only included those injuries with time loss ? 7 days. As a result, it is not surprising that our incidence proportions are lower than those found in the literature.   Acute-onset injuries accounted for 85.7% and 78.9% for the control and intervention cohorts respectively, similar to results reported in other studies (66.0-91.1% for control; 84.0-87.2% for intervention cohorts).(2-5,16) The lower proportion of acute-onset injuries in the intervention cohort may be a result of the small number of injuries that occurred in this cohort. The overall acute-onset incidence proportion in the control group was 13.7 injuries (95% CI, 8.5-22.1) per 100 players compared to 8.0 injuries (95% CI, 4.8-13.4) per 100 players in the intervention group. Acute-onset incidence proportions found in the literature are 19.5 to 22.2 and 11.1 to 19.7 acute-onset injuries per 100 players per season for control cohorts and intervention cohorts respectively.(2,3,16) Similar to LE injuries, our incidence proportion values for acute-onset injury are lower than those found in the literature. This is likely due to differences in the injury definitions used to estimate incidence proportion.   75 4.4.1.2 Age group  Consistent with previous studies, we found that injury risk was greater in the older age groups compared to the younger age groups.(13,16,26-28) Point estimates suggest that the incidence proportion for all injury, acute-onset injury and LE injury increases with increasing age group, independent of cohort. Our findings support the previous work of Emery & Meeuwisse who found an increased risk of all injury and acute-onset injury in the U16 to U18 group compared to the U13 to U15 group, independent of cohort.(16)   4.4.1.3 Injury location Consistent with the literature, ankles and knees were the most common body parts injured during both the control season and the intervention season.(2-5,12) During the control season, ankle injuries accounted for 30.4% of all injuries, consistent with previous prospective studies in youth female soccer players (23-35%).(2-5,12) During the intervention season, ankle injuries accounted for 15.8% of all injuries. Previous cluster randomized controlled trials have found that ankle injuries accounted for 32 to 37% of all injuries seen in the intervention cohorts.(2,3) Differences in the proportion of ankle injuries seen between our study and the literature may be a result of our injury definition as well as the small number of injuries that occurred. There were fewer ankle injuries in the intervention season compared to the control season suggesting that our injury prevention program was effective at reducing injuries to the ankle. Although not statistically significant, our exploratory analysis found that point estimates suggest the risk of ankle injury was reduced by 67% (95% CI, 0.10-1.07) and the risk of ankle sprain injury by 60% (95% CI, 0.12-1.35). The reduction in ankle sprain injury in our study agrees with results from a previous cluster RCT that found a reduction of 50% (95% CI, 0.24-1.04) between the control cohort and the intervention cohort.(16)  During the control season, knee injuries accounted for 16.1% of all injuries, consistent with other prospective studies in youth female soccer players (14-32%).(2-5,12) During the 2010-2011 season, knee injuries accounted for 31.6% of all injuries. Previous cluster RCTs have reported that knee injuries account for 18 to 22% of all injuries seen in the intervention groups.(2,3) When looking at knee sprain injury specifically, we found that the risk of injury was reduced by 58% (95% CI, 0.10-1.71) consistent with results in the literature.(16)  76 Differences in the proportion of knee injuries seen between our study and the literature may be a result of our injury definition, the type of injuries that occurred, or the small number of injuries that occurred.   Lower-extremity injuries accounted for 78.6% and 78.9% of all injuries in the control season and intervention season respectively. This is consistent with previous cohort studies that have found that LE injuries account for 67 to 100% of all injuries that occur in youth soccer players.(2-6,8-16) In a cluster RCT examining the effectiveness of a neuromuscular training program in reducing injury in youth soccer players, LE injuries accounted for 75.9% and 84.0% of all injuries seen in the control group and the training group respectively.(16)   4.4.1.4 Injury type Sprains and strains were the most common type of injury seen in both cohorts, consistent with the literature. During the control season, the most common types of injury were joint/ligament sprain (41.1%), muscle strain (28.6%) and fracture (7.1%). Prospective studies of youth female soccer players have found that the most common types of injuries are sprains (26.9-46.6%), strains (17.2-25.2%) and contusions (8.9-25.8%).(2-5,12) The high proportion of fracture injuries and low proportion of contusion injuries may be a result of the time loss injury definition used in our study.   During the intervention season, the most common types of injury were joint/ligament sprain (30.0%), muscle strain (20.0%), and contusion (15.0%). Previous intervention studies have found that sprains (42.2-57.1%) and strains (18.4-28.6%) comprised the majority of injuries seen in intervention groups.(2,3,12) The lower proportion of sprain injuries in the intervention group, compared to the literature, suggests that our injury prevention program may have a greater preventive effect on ligament sprains than those prevention programs previously studied. Discrepancies between our results and the literature, for the proportion of sprain injuries, may be a result of the injury definition used and the small number of injuries that occurred in the intervention season. Furthermore, two studies included only acute-onset injuries in their breakdown of injuries by type where our study included both acute-onset injuries and overuse injuries.(2,3)   77 Prospective cohort studies have shown that concussions account for 0.3 to 2.5% of all injuries that occur in youth female soccer players.(4,5) In our study, concussions accounted for 5.4% and 5.0% of the injuries that occurred during the 2008-2009 and 2010-2011 seasons respectively.   4.4.1.5 Injury severity Injury severity was classified according to the Consensus Statement definitions.(40) There did not appear to be any difference in the severity of injuries between the control cohort and the intervention cohort. It is difficult to compare our severity results to those in the literature as our definition only included injuries of time loss ? 7 days. The Consensus Statement definition for a mild injury includes any injury with time loss of 4 to 7 days.(40) Due to our time loss definition excluding mild injuries with time loss of 4 to 6 days, we are unable to compare our proportion of mild injuries to the literature. The proportion of injuries that were of moderate severity was 52.5% for the control group and 60.0% for the intervention group when only considering moderate and severe injuries. This is consistent with other intervention studies using similar definitions, which found that moderate severity injuries accounted for between 47.0% to 72.2% of all injuries (time loss > 7 days) in control cohorts and 56.0% to 92.9% of all injuries (time loss > 7 days) in the intervention groups.(2,3,16) The proportion of injuries that were classified as severe injuries was 47.5% for the control group and 40.0% for the intervention group when only considering moderate and severe injuries. Likewise, this is consistent with the literature, which found that severe injuries accounted for between 27.8% to 53.0% of all injuries (time loss > 7 days) in control cohorts and 7.1% to 44.0% of all injuries (time loss > 7 days) in the intervention groups.(2,3,16) Discrepancies between different studies may be due to the sample size and the low number of injuries that occurred in some of the studies.   4.4.1.6 Injury mechanism and session type  Contact was involved in 78.6% of all injuries in the control season, which is higher than a previous cohort study that found that 46.2% of all injuries involved contact.(6) Likewise, we found that contact was the mechanism of injury for 63.2% of all injuries in the intervention season, exceeding the values found in the literature (32.9-59.0%).(2,3,12) Consistent with the  78 literature, we found that a greater proportion of injuries occurred during games compared to practice.(3,6,12,34) Information about session type was missing for a large number of injuries (41.1%), which has the potential to greatly change the distribution of session type.   4.4.2 The effectiveness of the SIPP 4.4.2.1 Effect of intervention The unadjusted risk ratio point estimates suggest that this training program was effective in reducing all injury (RR, 0.64; 95% CI, 0.35-1.17), acute-onset injury (RR, 0.41; 95% CI, 0.29-1.17) and LE injury (RR, 0.64; 95% CI, 0.32-1.26) in the intervention cohort compared to the control group, however these results were not statistically significant. Using incidence proportion, we calculated the crude risk ratio for all injury, acute-onset injury, LE injury, and time loss injury (? 8 days) for the efficacy study by Emery & Meeuwisse.(16) Results from the Emery & Meeuwisse study suggest that the training program was efficacious at reducing all injury (RR, 0.61), acute-onset injury (RR, 0.56), LE injury (RR, 0.67) and time loss injury (RR, 0.76).(16) Risk ratio point estimates suggest that the magnitude of change was similar between our study and the previous efficacy study.(16) This suggests that a community-delivered neuromuscular team-training program may have a protective effect on all injury, acute-onset injury and LE injury.  4.4.2.2 Effect modification We hypothesized that there would not be any effect modification of age group on cohort based on the literature.(16) Our sub-group analysis suggests that the effect of the intervention on the risk of injury is different across age groups and that age may be an effect modifier in this study. Risk ratio point estimates suggest that the injury prevention program has the greatest effect on the reduction of all injury, acute-onset injury and LE injury in players in the older age groups. Further investigations of this interaction found that the relationship between age group and the intervention was not significant.   4.4.2.3 Effect of intervention- controlling for covariates The primary risk factor of interest was exposure to the injury prevention program. The secondary risk factors of interest were age group and level of play. The adjusted point  79 estimates suggest that this training program was effective in reducing all injury (RR, 0.73; 95% CI, 0.37-1.45), acute-onset injury (RR, 0.69; 95% CI, 0.33-1.44) and LE injury (RR, 0.74; 95% CI, 0.34-1.64) in the intervention cohort compared to the control group, however these results were not statistically significant. These analyses were adjusted for cluster by team and other covariates (age group and level of play). Results from our study agree with previous intervention studies that have also found a reduction in injury following the use of a neuromuscular training program.(2,8,12,16,33) One previous intervention study observed no difference in overall injury rates between youth female soccer players in the intervention and control groups.(3) It was thought that this was likely due to low compliance to the intervention.   Emery & Meeuwisse found significant reductions in the risk of all injury (IRR, 0.62; 95% CI, 0.39-0.99), acute-onset injury (IRR, 0.57; 95% CI, 0.35-0.91) and a non-significant reduction in the risk of LE injury (IRR, 0.68; 95% CI, 0.42-1.04) in the intervention group compared to the control group. (16) Similarly, Soligard et al. found a significant reduction in the risk of all injury (IRR, 0.68; 95% CI, 0.56-0.84) and acute-onset injury (IRR, 0.76; 95% CI, 0.61-0.95) in the group that participated in a comprehensive warm-up program compared to the control group.(2)   Consistent with the literature, adjusted point estimates suggest a greater risk of each of the three injury outcomes in the older age groups compared to the younger age groups.(13,16,26-28) Further, players in the more elite divisions of play were at increased risk of injury compared to players in the lowest division of play. This is consistent with a previous study in youth soccer that found that participation in the more elite divisions of play increased the risk of injury.(6)   Age group was not an effect modifier in our study but the addition of age group and level of play to the model increased the risk ratio for all injury from 0.64 (95% CI, 0.35-1.17) to 0.73 (95% CI, 0.37-1.45). The increase in the risk ratio point estimate suggests that either age group or level of play may be a confounder in our study but this relationship was not  80 significant. Independent of cohort, older players had a greater incidence of injury compared to younger players, which is consistent with the literature.(13,16,26-28)    4.4.3 Implementation context The community design of the study provided an opportunity to observe the implementation context surrounding the SIPP. The SIPP was implemented across all competitive female teams at the Semiahmoo Soccer Club. Injury prevention was an important topic at the club and members of the community were committed to preventing injuries in youth female soccer players. Implementation at the community level and delivery of the program by club affiliates promoted a culture of safety around the club. Community commitment and buy-in appeared to be necessary for adoption of the SIPP.   Likewise, coach buy-in seemed to play an important part in adherence to the SIPP. In a previous study in youth female soccer, coaches with positive attitudes towards injury prevention training were associated with teams that had high compliance to the FIFA ?11+.?(34) Frequent follow-up with teams was necessary to monitor adherence to the program and proper technique while performing the exercises. It is suggested that follow-up with teams should occur frequently and consistently until coaches and players are comfortable delivering the program.   Feedback from coaches and players was acted on in order to keep the teams participating in the SIPP. The Safety Director, trainer and Principal Investigator collaborated with coaches to revise the team component of the program according to coach and player feedback. Changes were made to increase the efficiency of the warm-up and to keep the players warm and engaged. Specifically, changes were made to the dynamic stretching and strength components of the program. Team designate feedback, on weekly exposure sheets, suggested that strength components were the most likely to be excluded from the warm-up program. This is likely due to the nature of the exercises as less active exercises that take place on the ground. Soccer in BC is played year-round which leads to sessions often taking place in cold and wet conditions. Weather and field conditions were cited as factors contributing to failure to complete the strength components of the program. Additionally, players found that the  81 strength exercises were not active enough and they tended to cool down through this phase of the program. Further research should investigate alternative, more active exercises to target strength training without having to be in direct contact with the ground. It is important to note that the soccer club and not the research team were responsible for initiating these changes to the prescribed injury prevention program.   4.4.4 Uptake of the SIPP Seven teams returned weekly exposure sheets with some completed adherence data. At the team level, adherence to the SIPP was 93% for games, 91% for practices and 92% overall. The Semiahmoo Soccer Club Safety Director and trainer led the injury prevention program twice weekly for a total of 10 weeks over the season. The evidence suggests that adherence to the SIPP was high although accurate adherence calculations were not possible.   4.4.5 Limitations We acknowledge that our study has several limitations.  4.4.5.1 Power The small sample size and the low number of injuries that occurred is a limitation of the current study. The sample size required to detect a difference of 50% between the control group and intervention group was 684 players (342 in each cohort). Our sample size was 538 (351 in the control group and 187 players in the intervention group) indicating that we did not have enough players in the intervention group to detect our necessary difference in proportions. Additionally, the small number of injuries limited our ability to determine the effect of the injury prevention program on specific injuries.   4.4.5.2 Selection bias During the 2010-2011 season, the inclusion rate for the club was low with only 15/20 (75%) teams and 187/323 (58%) of the eligible players participating in the study. Team recruitment began with the coaches and then proceeded to the individual players following team enrollment in the study. Five teams (n = 88 players) did not participate in the study due to coaches declining to participate. Teams and players that did not consent to participate in the  82 study may have differed from those that participated in the study in their attitudes towards injury prevention. Non-consenting players may have been less motivated to participate and comply with a prescribed injury prevention program. As a result, the non-participants may have had a higher incidence proportion than the study participants, which would have resulted in a higher study incidence proportion if these players had been included. In trying to assess the effectiveness of the injury prevention program when delivered by the community, exclusion of these players may contribute to an overestimation of the effect of the injury prevention program on the incidence of injury.  Another limitation of our study was that 96 players were included in both cohorts and we did not control for cluster by individual player. However, a sensitivity analysis was performed to determine if the make-up of the cohorts had an effect on the primary study outcomes. The risk ratio point estimates were found to be very similar in both analyses (with and without the 96 players included) leading us to conclude that our findings are robust to the overlap in players between the two cohorts and the players can be treated independently.   4.4.5.3 Measurement bias Injury rates are ideally expressed as the number of injuries that occur per 1,000 hours of athlete participation. Rates are expressed per 1,000 player-hours as a way to account for differences in exposure between players and teams throughout the study period. Weekly exposure data was not collected during the 2008-2009 season. Therefore, we were unable to include exposure hours as our offset variable in the Poisson Regression model and were only able to express injury incidence as the proportion of injuries that occurred per 100 players over one season. In order to compare incidence proportions between study cohorts, an assumption is made that participation hours between the 2008-2009 cohort and 2010-2011 cohort were similar.    Complete evaluation of the injury prevention program is restricted by the historical injury data collected. The historical data only addresses injuries of time loss ? 7 days. Therefore, only time loss injuries of ? 7 days can be used from the 2010-2011 season to estimate the risk ratio.   83 As with any multi-faceted program, the effect of individual components of the program on injury rates cannot be determined. Evaluation of the program is as a comprehensive entity only. Additionally, adherence data regarding team participation in the intervention was not comprehensive and even when available, the data collected did not tell us which components of the intervention the teams participated in. Adherence data for the home-based component of the program was not collected. Consequently, it is unknown how adherent the players were to the wobble board training. Adherence data for the wobble board training was not collected as poor return of self-report journals has been documented previously.(16) As a result, we are unable to assess the relationship between adherence to the home-based component of the program and injury rates.   4.4.5.4 Confounding We adjusted our model for variables that we hypothesized could be potential confounders (age group and level of play). History of injury has been found to be an independent risk factor for incident injury in youth soccer.(6,13,16,25) Unfortunately, this information was not available for the historical cohort so we were unable to control for previous injury in our models.   4.4.5.5 Generalizability The majority of the teams that did not participate in the study were older teams, which may have an effect on the external validity of the study. The data was collected from competitive female soccer players at the Semiahmoo Soccer Club in BC and therefore results may not be generalizable to other populations.   4.4.6 Strengths To our knowledge, this is the first study to assess the effectiveness of a community-driven neuromuscular training program at preventing time loss injuries in youth female soccer players. Revisions were made to the injury prevention program by volunteer coaches, trainers and the Safety Director at the club to improve adherence to the program. These changes were documented in Chapter 3. Despite the neuromuscular training program being slightly different than the one used in the intervention study by Emery & Meeuwisse(16), the results  84 from our study are consistent with those reported showing that a community-driven injury prevention program is protective against all injury, acute-onset injury and lower-extremity injury.   Previous intervention studies have only included youth 13 years and older in their studies.(2,3,8,12,16,33) Another strength of our study is that our cohort is younger than populations studied in the literature with inclusion of players as young as 9 years of age. This provides data on a population that has not been studied and may provide us with some guidance as to at what age injury prevention programs can and should be implemented.    4.5 Summary and conclusions Risk ratio point estimates suggest that a community-driven team-based neuromuscular training program reduces the risk of all injury, acute-onset injury and lower-extremity injury in youth female soccer players, however these findings were not statistically significant. The protective effect of such an injury prevention program initiated, taught, delivered and revised by volunteer coaches and trainers affiliated with the club is similar in magnitude to RCTs examining similar programs.   In conclusion, these findings highlight the need for future research on the implementation aspect of injury prevention programs to determine barriers and facilitators to adoption and sustainability of injury prevention strategies. Understanding the implementation context of injury prevention programs will allow for improved knowledge translation and movement of efficacy evidence into real-world settings.   85 Chapter  5: Risk factors for injury in youth female soccer players  5.1 Introduction Soccer is one of the most popular sports in Canada with approximately 328,000 youth female players registered in 2008.(1) In British Columbia (BC) alone there are 38,000 currently registered youth female players.(37) High rates of participation result in many youth players being at risk of sustaining a soccer-related injury. Injury incidence rates reported in youth female soccer are 1 to 5 injuries per 1,000 practice-hours and 8 to 22 injuries per 1,000 game-hours.(2-7) Incidence proportion in youth female soccer players ranges from 15 to 23 injuries per 100 players during games and 6 to 42 injuries per 100 players during practices.(2,3,5,6,12) The majority of injuries sustained by youth soccer players are to the lower-extremity (67- 100%).(2-6,8-16)   Stages 1 and 2 of the ?Translating Research into Injury Prevention Practice? (TRIPP) framework are necessary steps to build an evidence-based foundation upon which injury prevention programs can be developed.(82) The main objective of Stage 1 is to determine the burden of injury in the population through injury surveillance. TRIPP Stage 2 focuses on establishing the causes and mechanisms of injury as well as identifying risk and protective factors for injury.   Risk factors can be categorized as either intrinsic or extrinsic. Intrinsic risk factors are those that are internal to the athlete and include such risk factors as previous injury, sex, age, anthropometrics, fitness level, flexibility, strength, neuromuscular control, balance, and psychological factors. Extrinsic risk factors are those that act on the athlete externally and include such risk factors as playing surface, rules, level of play, experience, position and equipment.    Previous injury has consistently been shown to be a risk factor for future injury in youth soccer players.(6,13,16,25) There is also evidence that age group (5,6,13,16,26-29), skill level (6,30), joint laxity (31,32), and hamstring-to-quadriceps ratio (31) may be risk factors for injury in soccer. The objective of injury prevention programs is to target modifiable risk  86 factors to reduce the injury rate in the population of interest. The ?Soccer Injury Prevention Program? (SIPP) is designed to target strength, neuromuscular control and balance risk factors in order to prevent injury, particularly lower-extremity (LE) injury.(16) There are many effectiveness studies on injury prevention programs in youth soccer but these studies often fail to examine the risk factors when players have been exposed to an injury prevention program.   Therefore, this chapter is an exploratory analysis of all injury data collected during the 2010-2011 soccer season. Risk factors for all injury and lower-extremity injury are reported in youth female soccer players that were exposed to the SIPP. Risk factors of interest were determined a priori based on evidence from the literature.   5.2 Methods The recruitment process and general methods are outlined in chapter 3. All players from the intervention cohort (2010-2011) were included for analysis in this study. The injuries of interest were ?any soccer injury that results in the inability to complete a full session, and/or miss a subsequent session, and/or require medical attention.?(40)  Risk factors of interest were age group, level of play, previous injury, position, vertical drop jump (VDJ), sports participation, body mass index (BMI), aerobic fitness, unipedal balance, hip adduction-to-abduction ratio and knee flexion-to-extension ratio.   Operational risk factor definitions: 1. Categorical cut-offs were used to examine age group (U12, U14, U16, U18), level of play (Tier 1, 2, 3), and position (forward, midfield, defense, goalkeeper). Teams in the U13-U18 age categories played in Silver, Gold or Metro divisions while the U11 and U12 teams played in the Selects division. Players that played in the Metro or Selects 1 division were categorized as Tier 1. Players that played in the Gold or Selects 2 division were categorized as Tier 2. Players that played in the Silver division were categorized as Tier 3. Position was obtained through self-report on the preseason questionnaire (PQ).   87 2. A player was categorized as having a previous injury if they self-reported having an injury in the last year.  3. The mean of the sample was considered the cut-point for sports participation (hours/week), BMI (z-scores), aerobic fitness (mL/kg/min) and balance (seconds). Weekly sports participation was obtained from question 10 on the preseason questionnaire: ?Based on the past one year of activity, did you participate in any sport or combination of sports on a weekly basis??. 4. A player was considered to be high-risk for the vertical drop jump if they were rated as high-risk on at least one of the three trials. A second analysis was performed where players were considered high-risk for the vertical drop jump if they were rated as high-risk on all three trials.  5. Hip adduction-to-abduction strength ratio of 0.95 was used as the cut-point for the hip adduction-to-abduction strength ratio, as there is evidence that a ratio < 0.95 increases the risk of groin injuries in ice hockey players.(88)  6. Knee flexion-to-extension ratio of 0.5 was used as the cut-point for the knee flexion-to-extension risk factor, as there is evidence that a ratio < 0.5 increases the risk of injury in youth female soccer players.(73)  5.2.1 Statistical analysis 5.2.1.1 Descriptive epidemiology Descriptive statistics are provided for baseline characteristics of the cohort and for injured players versus uninjured players within the cohort. Categorical variables are presented as frequencies and percentages. Continuous variables are presented using means and 95% confidence intervals.   Descriptive statistics are used to describe characteristics of the injuries. Injury location, injury type, severity of injury, mechanism of injury, and session type when the injury occurred were described using frequencies and percentages. Definitions for severity were obtained from the Consensus Statement on Injury Definitions and Data Collection Procedures in Studies of Soccer Injuries.(40)    88 Injury rates (IR) for all injury, acute-onset injury and LE injury were estimated using Poisson Regression Analysis adjusted for cluster by team with player-hours included as the offset. Injury rates are presented as the number of injuries per 1,000 player-hours. Incidence proportion (IP) was estimated for all injury, acute-onset injury and LE injury using Poisson Regression Analysis adjusted for cluster by team. Incidence proportion is presented as the number of injuries that occurred within the study period per 100 players.   5.2.1.2 Incidence rate ratios Incidence rate ratios (IRR) were estimated to examine the association between potential risk factors and all injury and LE injury. IRR was estimated using univariate Poisson Regression Analysis adjusted for cluster by team with player-hours included as the offset.  For players that had no individual data and no team data available, imputation was based on the weekly sample means from players with completed data. When estimating the weekly sample mean, players contributed data if their weekly exposure was documented and they were not injured that week.   In the case that an individual player was missing all exposure hours but data was available for their teammates, hours were estimated using the weekly team mean. When an individual was missing some weeks of exposure data, the missing data was imputed based on the player?s individual means from the entire season as this best reflected individual variations in participation at practices and games.     Some weeks were exclusively estimated based on the sample mean as it was thought that the sample mean compared to the individual mean was more reflective of the true pattern of play during those weeks. This was the case for the weeks beginning September 5th (beginning of season), December 12th, 19th, and 26th (Winter break). We were unable to obtain the season end date for two teams, therefore we used the median number of weeks from the thirteen teams for which data was available.    89 A conservative approach was taken to estimate time loss for those players with missing exposure data. The number of games, practices, game-hours and practice-hours were multiplied by the proportion of the week missed to estimate time loss (e.g., for 2 days time loss, all values were multiplied by 2/7).   All tests were two-sided using a p-value of 0.05.  5.3 Results 5.3.1 Baseline characteristics There were 187 players in the study cohort. Figure 7 presents a flow chart outlining recruitment of players into the study and the number of players that participated in each component of baseline testing.    90                           Figure 7 Number of players who participated in the various components of baseline testing    Preseason questionnaire Balance Strength Aerobic fitness Vertical drop jump Anthropometrics n = 170 players n = 176 players  Hip ABD & ADD, n = 65 players Knee FLX, n = 49 players Knee EXT, n = 50 players   n = 113 players  n = 134 players  n = 177 players  Target population  Semiahmoo Soccer Club N =20 teams (n = 323 players)  Total number of participants recruited  N = 15 teams (n = 187 players)   91 Table 19 presents the baseline characteristics that were obtained through club roster information sheets and preseason questionnaires (PQ). We received completed PQs from 170 players (91%). Fifteen of the PQs (8.8%) were obtained retrospectively at the end of the season.   The mean age of the cohort was 13.1 years (95% CI, 12.9-13.4). There were fewer players in the U18 age group and the Tier 3 group compared to the other age and level of play categories respectively. Of the 170 players that returned the PQ, 59 players (31.6%) reported sustaining an injury in the previous year.   Table 19 Baseline characteristics, obtained through preseason questionnaire and official roster information  Characteristic All players n = 187 Median (range)a or Frequency (%) or  Mean (95% CI)b Age (years)a 13.4 (9.7-17.6) Age group  U12 46 (24.6) U14 65 (34.8) U16 68 (36.4) U18 8 (4.3) Level of play  Tier 1 61 (32.6) Tier 2 102 (54.5) Tier 3 24 (12.8) Position  Forward 43 (23.0) Midfield 61 (32.6) Defense 52 (27.8) Goal keeper 11 (5.9) Missing data  20 (10.7) Previous injury  Yes 59 (31.6) No 111 (59.4) Missing data  17 (9.1) Sports participation (hours/week)b 6.9 (6.4-7.4) Missing data 17 (9.1) CI, confidence interval  92 Time constraints and limited resources within each testing session restricted our ability to complete all components of the test battery for each subject that participated in baseline testing. This resulted in varying number of observations for the different tests (Figure 7). Participants who did not complete the Physical Activity Readiness Questionnaire (PAR-Q) or who had positive responses on the PAR-Q were not eligible to participate in the L?ger 20 m shuttle run test. A response was considered positive if the player answered yes to any of the seven questions on the PAR-Q. Twenty-six players had positive responses on the PAR-Q and eleven players failed to return the PAR-Q. There were 150 players eligible for participation in the 20 m shuttle run test of which 113 players completed the 20 m shuttle run test.  93 Table 20 presents baseline physical characteristics of the cohort collected during the baseline testing sessions.   Table 20 Baseline characteristics, physical attributes Characteristic All players n = 187 Frequency (%) or  Mean (95% CI) Height, cm 157.3 (155.9-158.7) Missing data (%) 10 (5.3) Weight, kg 48.8 (47.3-50.4) Missing data (%) 10 (5.3) BMI, kg/m2 19.5 (19.1-19.9) Missing data (%) 10 (5.3) Aerobic fitness, mL/kg/min 47.0 (46.3-47.7) Missing data (%) 74 (39.6) Maximum balance right, s 5.4 (4.9-5.9) Missing data (%) 10 (5.3) Maximum balance left, s 5.1 (4.6-5.5) Missing data (%) 11 (5.9) ADD:ABD ratio right 1.25 (1.17-1.34) Missing data (%) 122 (65.2) ADD:ABD ratio left 1.28 (1.18-1.38) Missing data (%) 122 (65.2) FLX:EXT ratio right 0.58 (0.53-0.63) Missing data (%) 138 (73.8) FLX:EXT ratio left 0.61 (0.56-0.67) Missing data (%) 138 (73.8) VDJ, high-risk rating if high-risk on any trial  High-risk 64 (34.2) Low-risk 76 (40.6) Missing data (%) 47 (25.1) VDJ, high-risk rating if high-risk on all 3 trials  High-risk 30 (16.0) Low-risk 110 (58.8) Missing data (%) 47 (25.1) CI, confidence interval; BMI, body mass index; ADD, adduction; ABD, abduction; FLX, flexion; EXT, extension; VDJ, vertical drop jump  94 Table 21 presents all of the baseline characteristics for the cohort by player injury status. These baseline characteristics will be further investigated to examine the risk factors for all injury and LE injury in youth female soccer players.   Table 21 Baseline characteristics by injury status Characteristic Injured players n= 34 Uninjured players n= 153 Mean (95% CI)   Age, years 12.8 (12.2-13.4) 13.2 (12.9-13.5) Height, cm 157.7 (153.8-161.6) 157.2 (155.7-158.7) Missing data (%) 4 (11.8) 6 (3.9) Weight, kg 49.0 (44.9-53.1) 48.8 (47.1-50.5) Missing data (%) 4 (11.8) 6 (3.9) BMI, kg/m2 19.5 (18.5-20.5) 19.5 (19.1-20.0) Missing data (%) 4 (11.8) 6 (3.9) Aerobic fitness, mL/kg/min 46.7 (44.9-48.5) 47.0 (46.3-47.8) Missing data (%) 13 (38.2) 61 (39.9) Max balance right, s 5.7 (4.4-7.0) 5.3 (4.8-5.8) Missing data (%) 4 (11.8) 6 (3.9) Max balance left, s 4.7 (4.1-5.3) 5.1 (4.6-5.7) Missing data (%) 4 (11.8) 7 (4.6) Sports participation, hours/week 6.5 (5.3-7.6) 7.0 (6.5-7.6) Missing data (%) 2 (5.9) 15 (9.8) ADD:ABD right 1.47 (1.23-1.71) 1.20 (1.12-1.28) Missing data (%) 21 (61.8) 101 (66.0) ADD:ABD left 1.41 (1.10-1.71) 1.25 (1.15-1.35) Missing data (%) 21 (61.8) 101 (66.0) FLX:EXT right 0.51 (0.42-0.61) 0.59 (0.54-0.65) Missing data (%) 26 (76.5) 112 (73.2) FLX:EXT left 0.59 (0.42-0.76) 0.62 (0.56-0.67) Missing data (%) 26 (76.5) 112 (73.2) No. (%)   Age group   U12 7 (20.6) 39 (25.5) U14 15 (44.1) 50 (32.7) U16 11 (32.4) 57 (37.3) U18 1 (2.9) 7 (4.6) Level of play   Tier 1 15 (44.1) 46 (30.1) Tier 2 17 (50.0) 85 (55.6) Tier 3 2 (5.9) 22 (14.4) Position   Forward 4 (11.8) 39 (25.5) Midfield 11 (32.4) 50 (32.7) Defense 13 (38.2) 39 (25.5) Goal keeper 3 (8.8) 8 (5.3) Missing data (%) 3 (8.8) 17 (11.1) Previous injury   Yes 14 (41.2) 45 (29.4) No 18 (52.9) 93 (60.8) Missing data (%) 2 (5.9) 15 (9.8)  95  Characteristic Injured players n= 34 Uninjured players n= 153 VDJ, high-risk rating if high-risk on any trial   High-risk 7 (20.6) 57 (37.3) Low-risk 18 (52.9) 58 (37.9) Missing data (%) 9 (26.5) 38 (24.8) VDJ, high-risk rating if high-risk on all 3 trials   High-risk 4 (11.8) 26 (17.0) Low-risk 21 (61.8) 89 (58.2) Missing data (%) 9 (26.5) 38 (24.8) CI, confidence interval; BMI, body mass index; ADD, adduction; ABD, abduction; FLX, flexion;  EXT, extension; VDJ, vertical drop jump  5.3.2 Descriptive epidemiology 5.3.2.1 Injury rates Table 22 provides a summary of the number of injuries, the injury rate and the incidence proportion for all injuries, acute-onset injuries and lower-extremity injuries.   During the 2010-2011 season 34 players sustained 41 injuries. Twenty-eight players sustained a single injury while five players sustained two independent injuries and one player sustained three independent injuries. The overall injury rate for all injury was 2.44 injuries per 1,000 player-hours (95% CI, 1.60-3.72). Of the 41 injuries, 35 (85.4%) were acute-onset injuries, 5 (12.2%) were overuse injuries and the injury type for one injury was unknown. The injury rate for acute-onset injuries was 2.08 acute-onset injuries per 1,000 player-hours (95% CI, 1.32-3.28). LE injuries accounted for 70.7% of all injuries resulting in an injury rate of 1.73 LE injuries per 1,000 player-hours (95% CI, 1.16-2.56).  There was low completion of weekly exposure sheets and 45.5% of total weeks were estimated based on completed data.   96 Table 22 Incidence proportion and injury rate for all injury, acute-onset injury and lower-extremity injury Outcome Athlete participation hours No. of injuries IP, injuries per 100 players (95% CI) IR, injuries per 1,000 player hours (95% CI) Overall     All injury 16,808 41 21.9 (13.9-34.6) 2.44 (1.60-3.72) Acute-onset injury 16,808 35 18.7 (11.4-30.9) 2.08 (1.32-3.28) LE injury 16,808 29 15.5 (10.4-23.1) 1.73 (1.16-2.56) Game     All injury 6,753 24 12.8 (7.4-22.2) 3.55 (2.19-5.75) Acute-onset injury 6,753 22 11.8 (6.5-21.2) 3.26 (1.95-5.44) LE injury 6,753 15 8.0 (5.2-12.4) 2.22 (1.50-3.30) Practice      All injury 10,055 15 8.0 (3.7-17.3) 1.49 (0.71-3.13) Acute-onset injury 10,055 13 7.0 (3.4-14.1) 1.29 (0.65-2.57) LE injury 10,055 13 7.0 (3.3-14.8) 1.29 (0.62-2.71) n =187 players; IP, incidence proportion; IR, injury rate; CI, confidence interval; LE, lower-extremity injury  97 5.3.2.2 Injury location and injury type  Table 23 provides a breakdown of all injuries by injury location and injury type. The most common locations of injury were the knee (24.4%), lower leg (14.6%), ankle (9.8%) and foot/toes (9.8%). The most common types of injury were contusion (25.0%), joint/ligament sprain (22.7%), and muscle strain (22.7%). Concussion accounted for 2.3% of all injuries that occurred during the season.   Table 23 Injury location and injury type  No (%) Body part  Knee 10 (24.4) Lower leg 6 (14.6) Ankle 4 (9.8) Foot/toes 4 (9.8) Upper leg 3 (7.3) Back 3 (7.3) Hip 2 (4.9) Wrist 2 (4.9) Finger 1 (2.4) Upper arm 1 (2.4) Head 1 (2.4) Neck 1 (2.4) Abdomen 1 (2.4) Ribs 1 (2.4) Other 1 (2.4) Injury typea  Contusion 11 (25.0) Joint/ligament sprain 10 (22.7) Muscle strain 10 (22.7) Fracture 1 (2.3) Concussion 1 (2.3) Dislocation 1 (2.3) Other 10 (22.7) a 41 injury events resulted in 44 types of injuries  98 5.3.2.3 Injury severity Table 24 presents severity of injury according to the definitions from the Consensus Statement.(40) During the 2010-2011 season, 8 injuries (19.5%) were classified as slight, 9 (22.0%) as minimal, 9 (22.0%) as mild, 9 (22.0%) as moderate and 6 (14.6%) as severe.   Table 24 Injury severity  No. (%) Slight: 0 days 8 (19.5) Minimal: 1-3 days 9 (22.0) Mild: 4-7 days 9 (22.0) Moderate: 8-28 days 9 (22.0) Severe: > 28 days 6 (14.6)  5.3.2.4 Injury mechanism and session type Table 25 provides information on the injury mechanism and session type when the injury occurred. During the 2010-2011 season, 54.8% of all injuries involved contact with another player while 40.5% of all injuries did not involve contact between players. The mechanism of injury was unknown for 2 injuries (4.8%). Throughout the season, 15 injuries (36.6%) occurred during practice, 24 (58.5%) during games, and the session type was unknown for 2 injuries (4.9%).   Table 25 Injury mechanism and session type  No. (%) Mechanisma  Contact 23 (54.8) Noncontact 17 (40.5) Unknown 2 (4.8) Session type  Game 24 (58.5) Practice 15 (36.6) Unknown 2 (4.9) aOne injury listed as both noncontact and unknown mechanism of injury  99 5.3.3 Risk factors for injury One hundred and seventy-seven players had BMI data available. International age-adjusted BMI cut-offs were first used to classify players into healthy weight, overweight or obese categories.(89) Of the players with BMI data available, 161 (91.0%) were considered healthy weight and 16 (9.0%) were considered overweight. No players were classified as obese. Since the distribution between the three categories was heavily skewed towards the healthy weight category these were not considered appropriate cut-offs to examine BMI as a potential risk factor for injury. Additionally, older girls tend to have a higher BMI than younger girls and this had the potential to confound our results if we used the mean BMI value as our cut-off for the risk factor. We transformed the BMI values to age-adjusted z-scores using the US Centers for Disease Control growth charts and then used the mean z-score value as the cut-point for risk of injury.(90)    5.3.3.1 All injury Table 26 summarizes IP, IR, and IRR estimates associated with potential risk factors for all injury. Age group, level of play, previous injury, vertical drop jump, sports participation, BMI, aerobic fitness, balance, and strength were not found to be risk factors for injury. Point estimates suggest that there was an increased risk of injury in U14 players compared to U12 players and a decreased risk of injury in U18 players compared to U12 players. Our study findings also suggest that the risk of injury is close to 2 times greater (IRR, 1.99; 95% CI, 0.32-12.40) for Tier 2 players and approximately 2.5 times greater (IRR, 2.59; 95% CI, 0.40-16.73) for Tier 1 players compared to Tier 3 players. The IRR point estimates suggest that having a knee flexion-to-extension ratio ? 0.5 increased the risk of injury but these findings were also not statistically significant.   There was evidence of player position being a significant risk factor for injury. There was an increased risk of injury for players who played the midfield (IRR, 2.25; 95% CI, 1.21-4.19) or defense positions (IRR, 2.99; 95% CI, 1.12-8.01) compared to players who played the forward position.     100 5.3.3.2 Lower-extremity injury  Lower-extremity (LE) injury was used as a primary outcome in this chapter since many of the baseline measurements are associated with LE injury.  Table 27 summarizes IP, IR, and IRR estimates associated with potential risk factors for LE injury. Age group, level of play, previous injury, vertical drop jump, sports participation, BMI, aerobic fitness, balance, adduction-to-abduction ratio, and left flexion-to-extension ratio were not found to be risk factors for LE injury. Point estimates suggest that there was a decreased risk of LE injury in U18 players compared to U12 players. The study findings suggest that risk of injury is close to 2 times greater (IRR, 1.98; 95% CI, 0.31-12.70) for Tier 1 players compared to Tier 3 players.   There was evidence that player position and right flexion-to-extension ratio were significant risk factors for LE injury. There was an increased risk of LE injury for players who played the defense position (IRR, 4.38; 95% CI, 1.08-17.75) compared to players who played the forward position. Point estimates suggest that there was also increased risk for players who played the midfield (IRR, 3.12; 95% CI, 0.81-12.02) and goalkeeper positions (IRR, 5.41; 95% CI, 0.82-35.85). Players that had a right knee flexion-to-extension ratio ? 0.5 had close to a 2.5 times (IRR, 2.43; 95% CI, 1.11-5.34) greater risk of LE injury compared to those players with a right knee flexion-to-extension ratio > 0.5.   101 Table 26 Risk factor analyses for all injury Risk factor Exposure hours No. of players No. of injuries IP, injuries/100 players (95% CI) IR, injuries/1,000 player-hours (95% CI) IRR (95% CI) Age group       U12 3,983 46 8 17.4 (5.9-51.6) 2.01 (0.67-6.05) 1 U14 6,323 65 19 29.2 (13.0-65.9) 3.00 (1.62-5.57) 1.50 (0.48-4.69) U16 5,632 68 13 19.1 (9.7-37.6) 2.31 (1.07-5.0) 1.15 (0.34-3.89) U18 870 8 1 12.5 (n/a) 1.15 (n/a) 0.57 (0.21-1.54) Level of play       Tier 3 1,711 24 2 8.3 (1.2-59.2) 1.17 (0.11-12.70) 1 Tier 2 9,472 102 22 21.6 (10.9-42.8) 2.32 (1.32-4.10) 1.99 (0.32-12.40) Tier 1 5,624 61 17 27.9 (14.7-53.0) 3.02 (1.48-6.17) 2.59 (0.40-16.73) Previous injury       No 10,014 111 21 18.9 (8.8-40.5) 2.09 (1.08-4.03) 1 Yes 5,213 59 16 27.1 (16.4-44.9) 3.06 (1.74-5.40) 1.47 (0.59-3.66) Position       Forward 3,775 43 4 9.3 (3.8-22.5) 1.05 (0.44-2.54) 1 Midfield 5,442 61 13 21.3 (9.8-46.6) 2.39 (1.19-4.80) 2.25 (1.21-4.19)* Defense 4,736 52 15 28.9 (19.7-42.2) 3.17 (1.99-5.03) 2.99 (1.12-8.01)* Goal keeper 1,047 11 4 36.4 (11.7-112.9) 3.82 (1.33-10.98) 3.61 (0.98-13.28) VDJ- high-risk rating if high-risk on any trial       Low-risk 7,116 76 23 30.3 (15.6-58.8) 3.23 (1.76-5.93) 1 High-risk 5,732 64 7 10.9 (4.2-28.6) 1.22 (0.48-3.11) 0.38 (0.12-1.20) VDJ- high-risk rating if high-risk on all 3 trials       Low-risk 10,072 110 26 23.6 (13.5-41.4) 2.58 (1.55-4.29) 1 High-risk 2,776 30 4 13.3 (4.3-41.2) 1.44 (0.49-4.22) 0.56 (0.20-1.59) Sports participationa       > 6.93 hours 6,491 72 12 16.7 (8.0-34.7) 1.85 (0.88-3.87) 1 ? 6.93 hours 8,766 98 25 25.5 (13.7-47.5) 2.85 (1.66-4.89) 1.54 (0.58-4.09)                                     102 Risk factor Exposure hours No. of players No. of injuries IP, injuries/100 players (95% CI) IR, injuries/1,000 player-hours (95% CI) IRR (95% CI) Body mass index (z-score)a       ? 0.12 7,236 78 17 21.8 (8.7-54.5) 2.35 (1.08-5.11) 1 > 0.12 8,727 99 20 20.2 (11.1-36.6) 2.29 (1.21-4.35) 0.98 (0.33-2.86) Aerobic fitnessa       > 47 mL 5,538 59 16 27.1 (13.3-55.5) 2.89 (1.60-5.23) 1 ? 47 mL 4,492 54 11 20.4 (10.6-39.0) 2.45 (1.30-4.60) 0.85 (0.38-1.89) Max balance righta       > 5.4 s 7,045 75 19 25.3 (13.5-47.4) 2.70 (1.65-4.40) 1 ? 5.4 s 8,992 102 18 17.7 (9.5-32.8) 2.00 (1.04-3.85) 0.74 (0.34-1.60) Max balance lefta       > 5.1 s 5,993 65 18 27.7 (16.9-45.3) 3.00 (1.92-4.71) 1 ? 5.1 s 9,958 111 19 17.1 (9.8-30.0) 1.91 (1.11-3.29) 0.64 (0.38-1.05) ADD:ABD right       > 0.95 5,027 53 17 32.1 (15.6-65.9) 3.38 (1.64-6.99) 1 ? 0.95 1,093 12 0 0 0 - ADD:ABD left       > 0.95 4,745 50 16 32.0 (15.0-68.5) 3.37 (1.55-7.36) 1 ? 0.95 1,374 15 1 6.7 (0.9-51.5) 0.73 (0.09-5.59) 0.22 (0.03-1.51) FLX:EXT right       > 0.5 2,943 31 5 16.1 (5.4-48.0) 1.70 (0.57-5.05) 1 ? 0.5 1,614 18 4 22.2 (6.9-72.1) 2.48 (0.65-9.40) 1.46 (0.58-3.64) FLX:EXT left       > 0.5 3,259 35 5 14.3 (4.8-43.0) 1.53 (0.48-4.88) 1 ? 0.5 1,298 14 4 28.6 (8.8-92.3) 3.08 (0.88-10.84) 2.01 (0.62-6.49) IP, incidence proportion; IR, injury rate; IRR, incidence rate ratio; CI, confidence interval; n/a, not available; VDJ, vertical drop jump ADD, adduction; ABD, abduction; FLX, flexion; EXT, extension;  a mean value for baseline measurement; * statistically significant at p < 0.05    103 Table 27 Risk factor analyses for lower-extremity injury Risk factor Exposure hours No. of players No. of injuries IP, injuries/100 players (95% CI) IR, injuries/1,000 player-hours (95% CI) IRR (95% CI) Age group       U12 3,983 46 7 15.2 (5.7-40.7) 1.76 (0.65-4.76) 1 U14 6,323 65 12 18.5 (9.2-37.2) 1.90 (1.05-3.44) 1.08 (0.38-3.08) U16 5,632 68 9 13.2 (6.7-26.0) 1.60 (0.72-3.55) 0.91 (0.28-2.90) U18 870 8 1 12.5 (n/a) 1.15 (n/a) 0.65 (0.27-1.60) Level of play       Tier 3 1,711 24 2 8.3 (1.2-59.2) 1.17 (0.11-12.70) 1 Tier 2 9,472 102 14 13.7 (7.9-23.8) 1.48 (0.91-2.40) 1.26 (0.21-7.71) Tier 1 5,624 61 13 21.3 (11.5-39.7) 2.31 (1.16-4.61) 1.98 (0.31-12.70) Previous injury       No 10,014 111 16 14.4 (8.7-24.0) 1.59 (1.02-2.48) 1 Yes 5,213 59 10 17.0 (10.2-28.2) 1.92 (1.09-3.39) 1.20 (0.60-2.44) Position       Forward 3,775 43 2 4.7 (1.3-16.6) 0.53 (0.15-1.89) 1 Midfield 5,442 61 9 14.8 (8.1-27.1) 1.65 (0.93-2.94) 3.12 (0.81-12.02) Defense 4,736 52 11 21.2 (13.7-32.7) 2.32 (1.42-3.79) 4.38 (1.08-17.75)* Goal keeper 1,047 11 3 27.3 (9.5-78.0) 2.87 (1.04-7.91) 5.41 (0.82-35.85) VDJ- high-risk rating if high-risk on any trial       Low-risk 7,116 76 16 21.1 (11.7-37.9) 2.25 (1.27-3.98) 1 High-risk 5,732 64 4 6.3 (1.9-20.9) 0.70 (0.21-2.31) 0.31 (0.07-1.32) VDJ- high-risk rating if high-risk on all 3 trials       Low-risk 10,072 110 118 16.4 (9.9-26.9) 1.79 (1.10-2.91) 1 High-risk 2,776 30 2 6.7 (1.7-25.5) 0.72 (0.19-2.67) 0.40 (0.10-1.67) Sports participationa       > 6.93 hours 6,491 72 9 12.5 (5.9-26.6) 1.39 (0.65-2.95) 1 ? 6.93 hours 8,766 98 17 17.4 (10.8-27.9) 1.94 (1.24-3.02) 1.40 (0.55-3.57)                              104  Risk factor Exposure hours # of players # of injuries IP, injuries/100 players (95% CI) IR, injuries/1,000 player-hours (95% CI) IRR (95% CI) Body mass index (z-score)a       ? 0.12 7,236 78 11 14.1 (7.3-27.4) 1.52 (0.87-2.67) 1 > 0.12 8,727 99 14 14.1 (7.6-26.4) 1.60 (0.82-3.13) 1.06 (0.42-2.67) Aerobic fitnessa       > 47 mL 5,538 59 9 15.3 (8.8-26.4) 1.63 (1.02-2.59) 1 ? 47 mL 4,492 54 8 14.8 (7.3-30.2) 1.78 (0.89-3.57) 1.10 (0.53-2.28) Max balance righta       > 5.4 s 7,045 75 14 18.7 (12.5-28.0) 1.99 (1.45-2.72) 1 ? 5.4 s 8,992 102 11 10.8 (5.3-21.8) 1.22 (0.58-2.57) 0.62 (0.28-1.36) Max balance lefta       > 5.1 s 5,993 65 13 20.0 (12.8-31.3) 2.17 (1.34-3.50) 1 ? 5.1 s 9,958 111 12 10.8 (6.3-18.6) 1.21 (0.70-2.07) 0.56 (0.29-1.05) ADD:ABD right       > 0.95 5,027 53 9 17.0 (7.6-38.1) 1.79 (0.74-4.33) 1 ? 0.95 1,093 12 0 - - - ADD:ABD left       > 0.95 4,745 50 8 16.0 (7.5-34.3) 1.69 (0.72-3.94) 1 ? 0.95 1,374 15 1 6.7 (0.9-51.5) 0.73 (0.09-5.59) 0.43 (0.08-2.32) FLX:EXT right       > 0.5 2,943 31 3 9.7 (4.1-22.9) 1.02 (0.42-2.44) 1 ? 0.5 1,614 18 4 22.2 (6.9-72.1) 2.48 (0.65-9.40) 2.43 (1.11-5.34)* FLX:EXT left       > 0.5 3,259 35 4 11.4 (4.0-33.1) 1.23 (0.40-3.81) 1 ? 0.5 1,298 14 3 21.4 (6.9-66.3) 2.31 (0.68-7.83) 1.88 (0.69-5.10) IP, incidence proportion; IR, injury rate; IRR, incidence rate ratio; CI, confidence interval; n/a, not available; VDJ, vertical drop jump ADD, adduction; ABD, abduction; FLX, flexion; EXT, extension;  a mean value for baseline measurement; * statistically significant at p < 0.05  105 5.4 Discussion The objective of this study was to identify risk factors for all injury and LE injury among players participating in a community-driven injury prevention program.   5.4.1 Descriptive epidemiology 5.4.1.1 Injury rates The cohort in this study was exposed to an injury prevention program throughout the season, therefore we compared our results to both cohorts that participated in an injury prevention program and cohorts that did not. The overall injury rate for the 2010-2011 season was 2.44 injuries per 1,000 player-hours (95% CI, 1.60-3.72). This is consistent with previous intervention studies that have found overall injury rates of 2.1 to 3.6 injuries per 1,000 player-hours in their intervention groups.(2,3,16) Overall, the reported injury incidence rates in youth female soccer ranges from 3.4 to 6.8 injuries per 1,000 player-hours.(2-6,16) Our observed injury rate is lower than the rates seen in prospective cohort studies where the players did not participate in an injury prevention program. This suggests that the injury prevention program used in the 2010-2011 season may have been effective in decreasing the rate of injury in youth female soccer players.   Game and practice injury rates in this study were 3.55 injuries per 1,000 game-hours (95% CI, 2.19-5.75) and 1.49 injuries per 1,000 practice-hours (95% CI, 0.71-3.13) respectively. Our observed game injury rate is lower than the game injury rates seen in other intervention cohorts (6.8-8.2 injuries/1,000 game-hours).(2,3) Similarly, the game injury rate in our study is lower than that observed in cohorts that did not participate in an injury prevention program (7.6-22.4 injuries/1,000 game-hours).(2-6) The observed practice injury rate is in agreement with injury rates seen in other intervention cohorts (0.9-1.5 injuries/1,000 practice-hours). (2,3) Likewise, our practice injury rate is consistent with injury rates previously reported in non-intervention cohorts (1.3-4.6 injuries per 1,000 practice-hours).(2-6) Reduced game injury rates in our study may be a result of greater adherence to the injury prevention program compared to previous studies, the younger age of the cohort, or the club culture around safety and injury prevention.    106 The acute-onset injury rate for this study was 2.08 acute-onset injuries (95% CI, 1.32-3.28) per 1,000 player-hours. This is consistent with results from two efficacy studies that reported acute-onset injury rates of 1.8 to 3.2 acute injuries per 1,000 player-hours.(3,16) The acute-onset injury rate in our study was lower than acute-onset injury rates reported in previous studies where the players did not participate in an injury prevention program (2.5-4.4 acute injuries per 1,000 player-hours).(3,4,16)  Lower-extremity injuries accounted for the majority of injuries in our study (70.7%), which resulted in an injury rate of 1.73 LE injuries (95% CI, 1.16-2.56) per 1,000 player-hours. Our reported LE injury rate is in agreement with the results from two efficacy studies that observed LE injury rates of 1.8 to 2.7 LE injuries per 1,000 player-hours in their intervention groups.(3,16) Our observed LE injury rate is lower than rates reported in two prospective cohort studies that did not involve participation in an injury prevention program (2.5-2.6 LE injuries/1,000 player-hours).(3,16) This finding suggests that our injury prevention program may have been effective in decreasing the rate of LE injury in youth female soccer players.   Our injury rates were generally in agreement with those observed in the training groups of previous intervention studies in youth female soccer. Differences between our results and the literature may be a result of the age of the study participants. The age range of our cohort was 9 to 17 years with a mean age of 13.1 years. The mean age of the cohorts in the literature were between 15 to 16 years of age (2-4) and the cohorts often only included players as young as 13 or 14 years of age.(2-4,16) There is evidence in the literature that injury risk is greater in older age groups compared to younger age groups.(13,16,26-28) Therefore, the fact that our cohort is younger than those seen in previous studies may explain why our injury rates are lower than those seen in the literature. There is also the possibility that injuries were missed in our study, which would lead to an underestimation of the injury rates.   5.4.1.2 Injury location and type Lower-extremity injuries accounted for 70.7% of all injuries that occurred in the 2010-2011 season. This is consistent with previous cohort studies that have found that LE injuries account for 67 to 100% of all injuries that occur in youth soccer players.(2-6,8-16) The knee  107 and lower leg were the most common body parts injured during the 2010-2011 season. Knee injuries accounted for 24.4% of all injuries, consistent with other prospective studies in youth female soccer players (14-32%).(2-5,12) Lower leg injuries accounted for 14.6% of all injuries that occurred in our cohort, which is consistent with results seen in the literature (5.0-15.4%).(2,4,5,12) Ankle injuries accounted for only 9.8% of all injuries, which is considerably lower than previous prospective studies in youth female soccer players (23-35%).(2-5,12) Previous cluster RCTs have found that ankle injuries accounted for 32 to 37% of all injuries seen in the intervention cohorts.(2,3) Differences in the proportion of ankle injuries between our study and the literature may be a result of the small number of injuries that occurred, missed injuries, the wobble board component of our intervention or use of other prophylactic measures that were not documented (e.g., ankle braces).   During the season, the most common types of injury were contusion (25.0%), joint/ligament sprain (22.7%) and muscle strain (22.7%). This is consistent with the literature where prospective studies in youth female soccer have reported that the most common types of injuries are sprains (26.9-46.6%), strains (17.2-25.2%) and contusions (8.9-25.8%).(2-5,12) The proportion of joint/ligament sprains reported in our study is slightly lower than the results seen in the literature. Discrepancies between our results and the literature may be a result of the injury definition used, the population studied, the small sample size, or the small number of injuries that occurred during the season. Concussion accounted for 2.3% of all injuries, which is consistent with previous studies (0.3-2.5%).(4,5)   5.4.1.3 Injury severity Injury severity was classified according to the Consensus Statement definitions.(40) Of the 41 injuries that occurred over the season, 33 resulted in time loss from soccer. There did not appear to be a difference in the proportion of injuries between injury severity categories. Injuries resulting in > 7 days time loss from soccer accounted for 36.6% of all injuries. It is difficult to compare injury severity results between our study and the literature as different injury definitions have been used and many studies only included injuries resulting in time loss from soccer. Our findings for the proportion of moderate and severe injury collectively are slightly higher than two studies that used the same definition to capture injuries (22.8- 108 25.7%).(6,16) Discrepancies between our results and the literature may be a result of the small sample size, small number of injuries, or an overrepresentation of injuries in the moderate and severe categories due to potential underreporting of injuries with short or no time loss.   5.4.1.4 Injury mechanism and session type Contact between players was involved in 54.8% of all injuries, which is slightly higher than previous cohort studies that found that 40.0% to 46.2% of all injuries involved contact.(6,9,11,42) Although our results differ slightly from the literature, the observed value for contact injuries falls within the 95% CI reported by Emery et al. (46.2%; 95% CI, 34.8-57.8%).(6) A limitation of this comparison is that our contact injury definition only included injuries that occurred as a result of contact with another player whereas the contact injury definition in the study by Emery et al. included injuries that occurred as a result of contact with another player or with equipment.   Consistent with the literature, we found that a greater proportion of injuries occurred during games compared to practices.(3,6,12,34) There was evidence that there was more than a twofold increase in risk of injury during game play than during practices (crude IRR, 2.38; 95% CI, 1.20-4.88). Likewise, point estimates suggest that there was close to a twofold increase in the risk of LE injury during game play compared to practices (crude IRR, 1.72; 95% CI, 0.76-3.92).   5.4.2 Risk factors for injury Age group, level of play, previous injury, vertical drop jump, sports participation, BMI, aerobic fitness, balance, and strength were not found to be risk factors for injury. There was evidence that player position was a significant risk factor for all injury. Age group, level of play, previous injury, vertical drop jump, sports participation, BMI, aerobic fitness, balance, adduction-to-abduction ratio, and left flexion-to-extension ratio were also not predictive of LE injury. However, there was evidence that player position and right flexion-to-extension ratio were significant risk factors for LE injury in this cohort.   109 5.4.2.1 Age group and level of play There was no evidence of age group being a risk factor for all injury or LE injury. Point estimates suggest that there may be an increased risk of injury in U14 players compared to U12 players and a decreased risk of injury in U18 players compared to U12 players. Similarly, the IRR point estimate suggests that there was a decreased risk of LE injury in the U18 players compared to the U12 players (IRR, 0.65; 95% CI, 0.27-1.60). The suggested decreased risk of injury in the U18 players may be a result of the small number of players in that group (n = 8) and the small number of injuries that the group sustained (n = 1). The majority of previous studies have found that injury risk is greater in older age groups compared to younger age groups in youth soccer.(13,16,26-29) In contrast, two studies found greater rates of injury in younger age groups compared to older age groups.(5,6)   IRR point estimates suggest that the risk of all injury and LE injury was increased in the Tier 1 and Tier 2 players compared to the Tier 3 players. Results in the literature are mixed with some studies finding an increased risk of injury in the more elite divisions of play (6) or more skilled players (30), and other studies finding an increased risk of injury in the lower divisions of play.(11,56) Our findings likely did not reach statistical significance due to the small sample size.   5.4.2.2 Position There was evidence of position of play being a significant risk factor for all injury and LE injury. There was an increased risk of all injury for players who played the midfield (IRR, 2.25; 95% CI, 1.21-4.19) or defense positions (IRR, 2.99; 95% CI, 1.12-8.01) compared to players who played the forward position. For LE injury, there was an increased risk for players who played the defense position only (IRR, 4.38; 95% CI, 1.08-17.75) compared to players who played the forward position. Results in the literature are conflicting with regards to position of play as a risk factor for injury. Similar to our results for all injury, a study in women?s professional soccer found that players who played the midfield position sustained a significantly higher number of injuries compared to the rest of the players.(57) Another study in elite women?s soccer reported that defenders and forwards were at a greater risk of injury than goalkeepers and midfielders.(54) Kucera et al. found that defenders were at significantly  110 greater risk for injury than midfield players in a prospective cohort study in youth soccer players.(13) Furthermore, Le Gall et al. (5) and Jacobson & Tegner (58) found that there was no significant difference in injury incidence rates between the various positions in youth and adult female players respectively. Differences in injury rates between positions have been thought to be due to the unique physical requirements of each position.    5.4.2.3 Previous injury, sports participation, BMI, aerobic fitness, and balance There was no evidence of previous injury, sports participation, BMI, aerobic fitness, and balance being risk factors for all injury or LE injury. Despite the fact that previous injury has been shown consistently to be a risk factor for injury in youth soccer(6,13,16,25), we found no significant association between previous injury and the risk of all injury or LE injury in our study. Low study power may have contributed to the lack of precision in the IRR estimate for previous injury.   It is difficult to assess BMI as a risk factor in this study population, as our group was relatively homogenous with the majority of players (91.0%) classified as healthy weight.(89) Kucera et al. reported that increasing BMI (using BMI quartiles) was associated with an increased risk of injury.(13) However, once adjusted for other covariates this relationship was no longer found to be significant. Emery et al. used the sample mean BMI value as the cut-point to assess the association between BMI and risk of injury.(6) Players with a higher BMI did not have an increased risk of injury compared to those players with a lower BMI.(6) Furthermore, studies in youth (25) and adult (32) female players also found no association between BMI and injury risk   The players in this study all played on competitive teams in the top three divisions of play in South Surrey and the Greater Vancouver area. VO2max estimates demonstrated that the players in our study had a high-level of aerobic fitness (mean, 47.0 ml/kg/min). Our findings are higher than normative data reported by L?ger et al. who found that the mean VO2max for 13-year-old girls was 44.4 ml/kg/min.(86) Classifying players using the Fitnessgram Healthy Zone Cut-offs reveals that 109 (99.1%) players fall into the healthy zone category for aerobic capacity.(91) Therefore, the group as a whole is relatively fit and the variation in aerobic  111 fitness may be too small to examine aerobic fitness as a risk factor in this population. Previous studies have reported no difference in estimated VO2max values between injured and uninjured players.(32) Only 113 players participated in the L?ger 20 m shuttle run test, which also may have limited our ability to find any evidence for VO2max as a risk factor for injury.   Our reported IRR estimate for balance agrees with a previous study by Emery et al. that demonstrated that dynamic balance was not associated with an increased risk of injury in youth soccer players.(6)  5.4.2.4 Vertical drop jump (VDJ) High-risk vertical drop jump was not predictive of all injury or LE injury. Contrary to our expected results, IRR point estimates suggest that a high-risk rating on the VDJ test was protective against all injury and LE injury. Hewett et al. observed that female athletes with increased dynamic knee valgus and abduction load (assessed using 3-D VDJ) were at increased risk of ACL injury.(23) Observational 2-D VDJ screening has been found to have acceptable rater agreement and specificity but inadequate sensitivity when compared to 3-D VDJ methods.(85) The use of 2-D VDJ screening may have been a limitation in evaluating this risk factor. Further, only 140 (74.9%) players participated in the VDJ test and the small sample size along with the small number of injuries may have been a limitation in our ability to accurately assess VDJ as a risk factor for injury   5.4.2.5 Strength There was an increased risk of injury in players that had a right knee flexion-to-extension ratio ? 0.5 compared to those players with a right knee flexion-to-extension ratio > 0.5. ?stenberg & Roos found no difference in isokinetic muscle strength when comparing injured players to uninjured players but they did not evaluate the ratio of the flexors to extensors.(32) Soderman et al. observed that a lower hamstring-to-quadriceps ratio (concentric) was associated with an increased risk of traumatic injuries.(31) In the same study, all of the players who sustained ACL injuries had a hamstring-to-quadriceps ratio of less than 0.55 on their injured side.(31) Contrary to the results for traumatic injuries, a higher ratio was found  112 to be a significant risk factor for overuse injury.(31) We did not analyze traumatic injuries and overuse injuries separately in our study due to the small number of injuries that occurred and the small number of players that had data for this baseline measurement. Injury risk was higher in female collegiate athletes with a knee flexion-to-extension ratio < 0.75 obtained through an isokinetic strength test at 180?/second.(66) Orchard et al. demonstrated an increased risk of hamstring injury in Australian Rules Football players with lower flexion-to-extension ratios obtained during an isokinetic strength test at 60?/second.(67)  There was no evidence of hip adduction-to-abduction ratio being a risk factor for all injury or LE injury. Tyler et al. found evidence that isometric hip adduction-to-abduction ratio may be a significant risk factor for predicting groin injury in male ice hockey players.(88) Due to the time-consuming nature of the test, only 65 (34.8%) players were able to participate in this baseline test.  5.4.3 Limitations We acknowledge that our study has several limitations.   5.4.3.1 Power The small sample size and the low number of injuries that occurred is a limitation of the current study. The primary objective of this thesis was to evaluate the effectiveness of the injury prevention program at reducing time loss injuries of ? 7 days in youth female soccer players. Sample size calculations were done a priori to find the necessary sample size to power the effectiveness study. This study chapter was exploratory in nature and was underpowered to predict significant risk factors for injury. Primary risk factors of interest were age group, level of play, previous injury and vertical drop jump. Exploratory analyses included evaluating BMI, position, sports participation, aerobic fitness, balance and strength risk factors.   The a priori sample size was calculated based on a predicted intracluster correlation of ? = 0.02. The intracluster correlation coefficient (?) calculated post hoc was ? = 0.08 resulting in an inflation factor of 1.88 compared to 1.22 that was used for the a priori sample size  113 calculation. Using the calculated intracluster correlation coefficient, we would require 526 players in each of the control and intervention group. The post hoc calculation of intracluster correlation coefficient suggests that our study was underpowered.   5.4.3.2 Selection bias During the 2010-2011 season, the inclusion rate for the club was low with only 15/20 (75%) teams and 187/323 (58%) of the eligible players participating in the study. Teams and players that did not consent to participate in the study may have differed from those that participated in the study in their attitudes towards injury prevention. Non-consenting players may have been less motivated to participate and comply with a prescribed injury prevention program. As a result, the non-participants may have had a higher injury rate than the study participants, which would have resulted in a higher study injury rate if these players had been included.   5.4.3.3 Measurement bias Weekly exposure hours were estimated for some teams and players during the 2010-2011 season as a result of low completion of weekly exposure sheets. Five teams (33.3%) did not return any weekly exposure sheets and data had to be estimated for all weeks for these teams. Overall, 45.5% of total weeks were estimated.   Multiple comparisons may have increased the likelihood of a type I error being made therefore caution should be taken when interpreting the significance of the results. The analyses were primarily performed to provide some evidence to inform future research.   Furthermore, baseline testing was not completed on all players due to lack of time and resources. Testing was only done at the beginning of the season and we are unable to assess how a player?s individual risk factors may have changed over the season. Risk factors that act on a player are dynamic and it is important to assess them throughout the season whenever possible.     114 5.4.3.4 Confounding The analyses in our risk factor study were univariate and we did not control for possible confounders.   5.4.3.5 Generalizability The majority of the teams that did not participate in the study were older teams, which may have an effect on the external validity of the study. The data was collected from competitive female soccer players at the Semiahmoo Soccer Club in BC and therefore results may not be generalizable to other populations.   5.4.4 Strengths  This study provides information on potential risk factors for all injury and LE injury in youth female soccer players exposed to an injury prevention program. All of the players is this study were exposed to the injury prevention program consistently throughout the season. Results of this study may provide evidence to support modification of the SIPP, or development of additional prevention programs, to further reduce injuries in youth female soccer players.  5.5 Summary and conclusions Position of play and right knee flexion-to-extension ratio were found to be significant risk factors for injury in youth female soccer players participating in an injury prevention program. The risk of all injury was greater in defenders and midfielders and the risk of lower-extremity injury was greater in defenders compared to forwards. Players with a right knee flexion-to-extension ratio ? 0.5 had close to a 2.5 times greater risk of lower-extremity injury compared to those players with a ratio > 0.5.   In conclusion, our small sample size was likely the main limitation in examining associations between potential risk factors and injury. Results from this study should be used to inform future research.   115 Chapter  6: Summary and conclusions  The overall aim of this thesis was to determine the relationship between injury, risk factors and a community-driven injury prevention program in youth female soccer players. Chapter 4 includes the results of a historical cohort study evaluating the effectiveness of a community-driven neuromuscular training program in reducing soccer-related injuries in youth female soccer players. This is an important question given the high rates of injury in this population. The key finding of this chapter is that risk ratio point estimates suggest that a community-driven team-based neuromuscular training program reduces the risk of all injury, acute-onset injury and lower-extremity injury in youth female soccer players. The protective effect of such an injury prevention program initiated, taught, delivered and revised by volunteer coaches and trainers affiliated with the club is similar in magnitude to RCTs examining similar programs.(2,16) These findings highlight the need for future research focused on the effectiveness and implementation stages of injury prevention research. In future, studies should examine the implementation context including barriers and facilitators to adoption of the SIPP.   Chapter 5 examined the risk factors for youth female soccer players participating in a neuromuscular training warm-up program. Position of play (defense and midfield compared to forwards) and right knee flexion-to-extension ratio (? 0.5) were found to be significant risk factors for injury in youth female soccer players exposed to an injury prevention program. The small sample size was a limitation in examining associations between potential risk factors and injury in this exploratory analysis. In future, studies with a larger sample size should be done to examine risk factors, particularly modifiable risk factors, for players who are exposed to an injury prevention program. Further, measurement of risk factors should take place at more than one point in the season to assess how the athlete risk profile changes over time. Future research should continue to investigate risk factors to better understand the risk factors associated with injury in youth female soccer players in order to modify injury prevention programs to improve their effectiveness.   116 6.1 Recommendations for implementation of the SIPP 1) Implementation of the SIPP should take place at the community level with all teams affiliated with the club participating.   2) Implementation of a neuromuscular training program should be led by a team of highly committed individuals. Follow-up with teams should be constant until the coaches and players are comfortable delivering the program.    3) Coach buy-in is essential for successful implementation of an injury prevention program.   4) Injury prevention programs may be successfully implemented in youth as young as 9 years of age.   5) Feedback from coaches and players should be taken into consideration in order to maintain high levels of compliance. Trained personnel should collaborate with coaches to revise the program where necessary while still preserving the integrity of the program and as many components as possible.   In conclusion, community-driven injury prevention suggests a protective effect similar in magnitude to RCTs examining similar programs. Addressing the gaps in injury prevention research will facilitate the development of effective injury prevention programs with high uptake and compliance. This in turn will reduce the rate of injury in youth soccer players allowing youth to continue to participate in sport and reduce the potential for long-term negative consequences of injury.   117 References (1) Canadian Soccer Association. 2008 CSA demographics report. Available at: http://www.canadasoccer.com/files/2008_demographics_report.pdf. Accessed December 3, 2010.  (2) Soligard T, Myklebust G, Steffen K, Holme I, Silvers H, Bizzini M, et al. Comprehensive warm-up programme to prevent injuries in young female footballers: cluster randomised controlled trial. Br Med J. 2008;337:a2469.  (3) Steffen K, Myklebust G, Olsen OE, Holme I, Bahr R. Preventing injuries in female youth football--a cluster-randomized controlled trial. Scand J Med Sci Sports. 2008 Oct;18(5):605-614.  (4) Soderman K, Adolphson J, Lorentzon R, Alfredson H. Injuries in adolescent female players in European football: a prospective study over one outdoor soccer season. Scand J Med Sci Sports. 2001 Oct;11(5):299-304.  (5) Le Gall F, Carling C, Reilly T. Injuries in young elite female soccer players. Am J Sports Med. 2008;36(2):276-284.  (6) Emery CA, Meeuwisse WH, Hartmann SE. Evaluation of risk factors for injury in adolescent soccer: implementation and validation of an injury surveillance system. Am J Sports Med. 2005 Dec;33(12):1882-1891.  (7) Schiff MA, Mack CD, Polissar NL, Levy MR, Dow SP, O'Kane JW. Soccer injuries in female youth players: comparison of injury surveillance by certified athletic trainers and internet. J Athl Train. 2010;45(3):238-242.  (8) Junge A, Rosch D, Peterson L, Graf-Baumann T, Dvorak J. Prevention of soccer injuries: a prospective intervention study in youth amateur players. Am J Sports Med. 2002 Sep-Oct;30(5):652-659.  (9) Kakavelakis KN, Vlazakis S, Vlahakis I, Charissis G. Soccer injuries in childhood. Scand J Med Sci Sports. 2003 Jun;13(3):175-178.  (10) Brito J, Rebelo A, Soares JM, Seabra A, Krustrup P, Malina RM. Injuries in youth soccer during the preseason. Clin J Sport Med. 2011;21(3):259-260.  (11) Peterson L, Junge A, Chomiak J, Graf-Baumann T, Dvorak J. Incidence of football injuries and complaints in different age groups and skill-level groups. Am J Sports Med. 2000;28(suppl):S51-7.  (12) Heidt RS Jr, Sweeterman LM, Carlonas RL, Traub JA, Tekulve FX. Avoidance of soccer injuries with preseason conditioning. Am J Sports Med. 2000 Sep-Oct;28(5):659-662.   118 (13) Kucera K, Marshall S, Kirkendall D, Marchak P, Garrett Jr W. Injury history as a risk factor for incident injury in youth soccer. Br J Sports Med. 2005;39(7):462-466.  (14) Emery CA, Meeuwisse WH. Risk factors for injury in indoor compared with outdoor adolescent soccer. Am J Sports Med. 2006 Oct;34(10):1636-1642.  (15) Le Gall F, Carling C, Reilly T, Vandewalle H, Church J, Rochcongar P. Incidence of injuries in elite French youth soccer players. Am J Sports Med. 2006;34(6):928-938.  (16) Emery CA, Meeuwisse WH. The effectiveness of a neuromuscular prevention strategy to reduce injuries in youth soccer: a cluster-randomised controlled trial. Br J Sports Med. 2010 Jun;44(8):555-562.  (17) Grimmer KA, Jones D, Williams J. Prevalence of adolescent injury from recreational exercise: an Australian perspective. J Adolesc Health. 2000;27(4):266-272.  (18) Myklebust G, Bahr R. Return to play guidelines after anterior cruciate ligament surgery. Br J Sports Med. 2005;39(3):127-131.  (19) Arendt E, Dick R. Knee injury patterns among men and women in collegiate basketball and soccer: NCAA data and review of literature. Am J Sports Med. 1995 Nov-Dec;23(6):694-701.  (20) Ferretti A, Papandrea P, Conteduca F, Mariani PP. Knee ligament injuries in volleyball players. Am J Sports Med. 1992;20(2):203-207.  (21) Gray J, Taunton J, McKenzie D, Clement D, McConkey J, Davidson R. A survey of injuries to the anterior cruciate ligament of the knee in female basketball players. Int J Sports Med. 1985;6(6):314-316.  (22) Myklebust G, Maehlum S, Holm I, Bahr R. A prospective cohort study of anterior cruciate ligament injuries in elite Norwegian team handball. Scand J Med Sci Sports 1998;8(3):149-153.  (23) Hewett TE, Myer GD, Ford KR, Heidt RS Jr, Colosimo AJ, McLean SG, et al. Biomechanical measures of neuromuscular control and valgus loading of the knee predict anterior cruciate ligament injury risk in female athletes. Am J Sports Med. 2005;33(4):492-501.  (24) Ford KR, Myer GD, Toms HE, Hewett TE. Gender differences in the kinematics of unanticipated cutting in young athletes. Med Sci Sports Exerc. 2005;37(1):124-129.  (25) Steffen K, Myklebust G, Andersen TE, Holme I, Bahr R. Self-reported injury history and lower limb function as risk factors for injuries in female youth soccer. Am J Sports Med. 2008 Apr;36(4):700-708.  119 (26) Schmidt-Olsen S, B?nemann L, Lade V, Brass?e J. Soccer injuries of youth. Br J Sports Med. 1985;19(3):161-164.  (27) Sullivan JA, Gross RH, Grana WA, Garcia-Moral CA. Evaluation of injuries in youth soccer. Am J Sports Med. 1980;8(5):325-327.  (28) Yde J, Nielsen AB. Sports injuries in adolescents' ball games: soccer, handball and basketball. Br J Sports Med. 1990 Mar;24(1):51-54.  (29) Nielsen AB, Yde J. Epidemiology and traumatology of injuries in soccer. Am J Sports Med. 1989;17(6):803-807.  (30) Soligard T, Grindem H, Bahr R, Andersen TE. Are skilled players at greater risk of injury in female youth football? Br J Sports Med. 2010;44(15):1118-1123.  (31) Soderman K, Alfredson H, Pietila T, Werner S. Risk factors for leg injuries in female soccer players: a prospective investigation during one out-door season. Knee Surg Sports Traumatol Arthrosc. 2001 Sep;9(5):313-321.  (32) ?stenberg A, Roos H. Injury risk factors in female European football. A prospective study of 123 players during one season. Scand J Med Sci Sports. 2000 Oct;10(5):279-285.  (33) Mandelbaum BR, Silvers HJ, Watanabe DS, Knarr JF, Thomas SD, Griffin LY, et al. Effectiveness of a neuromuscular and proprioceptive training program in preventing anterior cruciate ligament injuries in female athletes: 2-year follow-up. Am J Sports Med. 2005 Jul;33(7):1003-1010.  (34) Soligard T, Nilstad A, Steffen K, Myklebust G, Holme I, Dvorak J, et al. Compliance with a comprehensive warm-up programme to prevent injuries in youth football. Br J Sports Med. 2010;44(11):787-793.  (35) FIFA. FIFA big count 2006: 270 million people active in football. Available at: http://www.fifa.com/mm/document/fifafacts/bcoffsurv/bigcount.statspackage%5f7024.pdf. Accessed December 3, 2010.  (36) Ifedi F. Sport participation in Canada, 2005. Ottawa, ON: Statistics Canada, 2008.  (37) Stevens R. Youth female soccer participation in BC. Email from rebeccastevens@bcsoccer.net. Received May 25, 2012.  (38) Scime G, Rajabali F, Turcotte K, Jivani K, Babul S, Pike I. Children and youth in British Columbia- injury by ages and stages, 2003-2007. Vancouver, BC: BC Injury Research and Prevention Unit, 2010 Nov.  (39) Van Mechelen W, Hlobil H, Kemper H. Incidence, severity, aetiology and prevention of sports injuries. A review of concepts. Sports Med. 1992;14(2):82-99.  120 (40) Fuller CW, Ekstrand J, Junge A, Andersen TE, Bahr R, Dvorak J, et al. Consensus statement on injury definitions and data collection procedures in studies of football (soccer) injuries. Clin J Sport Med. 2006 Mar;16(2):97-106.  (41) Nilsson S, Roaas A. Soccer injuries in adolescents. Am J Sports Med. 1978;6(6):358-361.  (42) Yard EE, Schroeder MJ, Fields SK, Collins CL, Comstock RD. The epidemiology of United States high school soccer injuries, 2005?2007. Am J Sports Med. 2008;36(10):1930-1937.  (43) Elias SR. 10-year trend in USA Cup soccer injuries: 1988-1997. Med Sci Sports Exerc. 2001 Mar;33(3):359-367.  (44) Williamson J, Rice SG. Incidence of injury in a large youth soccer tournament. Clin J Sport Med. 2006;16(5):435.  (45) Soligard T, Bahr R, Andersen T. Injury risk on artificial turf and grass in youth tournament football. Scand J Med Sci Sports. 2012 Jun;22(3):356-361.  (46) Kiani A, Hellquist E, Ahlqvist K, Gedeborg R, Micha?lsson K, Byberg L. Prevention of soccer-related knee injuries in teenaged girls. Arch Intern Med. 2010;170(1):43-49.  (47) Wald?n M, Atroshi I, Magnusson H, Wagner P, H?gglund M. Prevention of acute knee injuries in adolescent female football players: cluster randomised controlled trial. Br Med J. 2012 May;344:e3042.  (48) FACSM D, Marchak P, FACSM W. A prospective 3-year injury incidence in youth soccer. Med Sci Sports Exerc. 2002;34(suppl):S101.  (49) Kibler WB. Injuries in adolescent and preadolescent soccer players. Med Sci Sports Exerc. 1993 Dec;25(12):1330-1332.  (50) Meeuwisse WH. Assessing causation in sport injury: a multifactorial model. Clin J Sport Med. 1994;4(3):166-170.  (51) Meeuwisse WH, Tyreman H, Hagel B, Emery C. A dynamic model of etiology in sport injury: the recursive nature of risk and causation. Clin J Sport Med. 2007;17(3):215-219.  (52) Arnason A, Sigurdsson SB, Gudmundsson A, Holme I, Engebretsen L, Bahr R. Risk factors for injuries in football. Am J Sports Med. 2004;32(suppl):S5-S16.  (53) H?gglund M, Wald?n M, Ekstrand J. Previous injury as a risk factor for injury in elite football: a prospective study over two consecutive seasons. Br J Sports Med. 2006;40(9):767-772.   121 (54) Faude O, Junge A, Kindermann W, Dvorak J. Risk factors for injuries in elite female soccer players. Br J Sports Med. 2006 Sep;40(9):785-790.  (55) Henderson G, Barnes CA, Portas MD. Factors associated with increased propensity for hamstring injury in English Premier League soccer players. J Sci Med Sport. 2010;13(4):397-402.  (56) Junge A, Chomiak J, Dvorak J. Incidence of football injuries in youth players. Am J Sports Med. 2000;28(suppl):S47-S50.  (57) Giza E, Mith?fer K, Farrell L, Zarins B, Gill T. Injuries in women?s professional soccer. Br J Sports Med. 2005;39(4):212-216.  (58) Jacobson I, Tegner Y. Injuries among female football players-With special emphasis on regional differences. Adv Physiother. 2006;8(2):66-74.  (59) Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Intrinsic risk factors for acute ankle injuries among male soccer players: a prospective cohort study. Scand J Med Sci Sports. 2010 Jun;20(3):403-410.  (60) Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Intrinsic risk factors for acute knee injuries among male football players: a prospective cohort study. Scand J Med Sci Sports. 2011 Oct;21(5):645-652.  (61) Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Intrinsic risk factors for hamstring injuries among male soccer players: a prospective cohort study. Am J Sports Med. 2010 Jun;38(6):1147-1153.  (62) Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Intrinsic risk factors for groin injuries among male soccer players: a prospective cohort study. Am J Sports Med. 2010 Oct;38(10):2051-2057.  (63) Frisch A, Urhausen A, Seil R, Croisier JL, Windal T, Theisen D. Association between preseason functional tests and injuries in youth football: A prospective follow?up. Scand J Med Sci Sports. 2011 Dec;21(6):e468.  (64) Fousekis K, Tsepis E, Poulmedis P, Athanasopoulos S, Vagenas G. Intrinsic risk factors of non-contact quadriceps and hamstring strains in soccer: a prospective study of 100 professional players. Br J Sports Med. 2011 Jul;45(9):709-714.  (65) Hewett TE, Lindenfeld TN, Riccobene JV, Noyes FR. The effect of neuromuscular training on the incidence of knee injury in female athletes. A prospective study. Am J Sports Med. 1999 Nov-Dec;27(6):699-706.    122 (66) Knapik JJ, Bauman CL, Jones BH, Harris JMA, Vaughan L. Preseason strength and flexibility imbalances associated with athletic injuries in female collegiate athletes. Am J Sports Med. 1991;19(1):76-81.  (67) Orchard J, Marsden J, Lord S, Garlick D. Preseason hamstring muscle weakness associated with hamstring muscle injury in Australian footballers. Am J Sports Med. 1997;25(1):81-85.  (68) Hewett TE, Stroupe AL, Nance TA, Noyes FR. Plyometric training in female athletes. Decreased impact forces and increased hamstring torques. Am J Sports Med. 1996 Nov-Dec;24(6):765-773.  (69) Myer GD, Ford KR, McLean SG, Hewett TE. The effects of plyometric versus dynamic stabilization and balance training on lower extremity biomechanics. Am J Sports Med. 2006 Mar;34(3):445-455.  (70) Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Prevention of injuries among male soccer players: a prospective, randomized intervention study targeting players with previous injuries or reduced function. Am J Sports Med. 2008 Jun;36(6):1052-1060.  (71) Caraffa A, Cerulli G, Projetti M, Aisa G, Rizzo A. Prevention of anterior cruciate ligament injuries in soccer. A prospective controlled study of proprioceptive training. Knee Surg Sports Traumatol Arthrosc. 1996;4(1):19-21.  (72) Myer GD, Ford KR, Paterno MV, Nick TG, Hewett TE. The effects of generalized joint laxity on risk of anterior cruciate ligament injury in young female athletes. Am J Sports Med. 2008 Jun;36(6):1073-1080.  (73) Allen SPW. Do Components of a Pre-participation Examination in Female Adolescent Soccer Players Predict Acute Lower Extremity Injuries in Soccer? [MSc thesis]. Calgary, AB: University of Calgary; 2011.  (74) Steffen K, Andersen TE, Bahr R. Risk of injury on artificial turf and natural grass in young female football players. Br J Sports Med. 2007 Aug;41(suppl):i33-i37.  (75) Aoki H, Kohno T, Fujiya H, Kato H, Yatabe K, Morikawa T, et al. Incidence of injury among adolescent soccer players: a comparative study of artificial and natural grass turfs. Clin J Sport Med. 2010 Jan;20(1):1-7.  (76) Ekstrand J, H?gglund M, Fuller C. Comparison of injuries sustained on artificial turf and grass by male and female elite football players. Scand J Med Sci Sports. 2011 Dec;21(6):824-832.  (77) Ekstrand J, Timpka T, Hagglund M. Risk of injury in elite football played on artificial turf versus natural grass: a prospective two-cohort study. Br J Sports Med. 2006 Dec;40(12):975-980.  123 (78) Fuller CW, Dick RW, Corlette J, Schmalz R. Comparison of the incidence, nature and cause of injuries sustained on grass and new generation artificial turf by male and female football players. Part 2: training injuries. Br J Sports Med. 2007 Aug;41(suppl):i27-i32.  (79) Fuller CW, Dick RW, Corlette J, Schmalz R. Comparison of the incidence, nature and cause of injuries sustained on grass and new generation artificial turf by male and female football players. Part 1: match injuries. Br J Sports Med. 2007 Aug;41(suppl):i20-i26.  (80) Bahr R, Engebretsen L. Sports injury prevention. Wiley-Blackwell; 2009.  (81) Emery CA. Considering cluster analysis in sport medicine and injury prevention research. Clin J Sport Med. 2007 May;17(3):211-214.  (82) Finch C. A new framework for research leading to sports injury prevention. J Sci Med Sport. 2006;9(1):3-9.  (83) Donner A, Klar N. Design and analysis of cluster randomization trials in health research. London: Arnold; 2000.  (84) Emery CA, Cassidy JD, Klassen TP, Rosychuk RJ, Rowe BB. Development of a clinical static and dynamic standing balance measurement tool appropriate for use in adolescents. Phys Ther. 2005 Jun;85(6):502-514.  (85) Ekegren CL, Miller WC, Celebrini RG, Eng JJ, MacIntyre DL. Reliability and validity of observational risk screening in evaluating dynamic knee valgus. J Orthop Sports Phys Ther. 2009;39(9):665-674.  (86) L?ger LA, Mercier D, Gadoury C, Lambert J. The multistage 20 metre shuttle run test for aerobic fitness. J Sports Sci. 1988;6(2):93-101.  (87) Leetun DT, Ireland ML, Willson JD, Ballantyne BT, Davis I. Core stability measures as risk factors for lower extremity injury in athletes. Med Sci Sports Exerc. 2004;36(6):926-934.  (88) Tyler TF, Nicholas SJ, Campbell RJ, McHugh MP. The association of hip strength and flexibility with the incidence of adductor muscle strains in professional ice hockey players. Am J Sports Med. 2001;29(2):124-128.  (89) Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Br Med J. 2000;320(7244):1240-1245.  (90) Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, et al. CDC growth charts: United States. Adv Data. 2000(314):1.   124 (91) The Cooper Institute. Fitnessgram & Activitygram Test Administration Manual-Updated 4th Edition. Human Kinetics; 2010.  125  Appendices Appendix A Sample Size Calculations ? = 0.05 = acceptable type I error (using 2-tailed test) ? = 0.20 = acceptable type II error pc = 0.15 = proportion of adolescents in the control group anticipated to sustain a ? 1week time loss injury during the 6 month study period (based on injuries ? 1week time loss, U14-U18 in previous RCT control group) (16)  pI = 0.075 = proportion of adolescents in the intervention group anticipated to sustain a ? 1week time loss injury during the 6 month study period pm = 0.1125 = the mean of pc and pI ?= 0.02 = estimated intracluster correlation coefficient (based on RCT soccer study) (16)  c = adjusted n/m = number of teams (clusters) in each group  The number of adolescents required per group is calculated by: N = 2 (z1-?/2 + z1-?)2(pm)(1-pm)/ (pc ? pI)2       = 2 (1.96 + 0.85)2(0.1125)(0.8875)/ (0.15 ? 0.075)2     = 1.57675/ 0.005625 = 280   m = average cluster size  (the number of adolescents per team required based on 23 participating teams in control year) and the above sample size calculation prior to adjusting for cluster randomization = 12  The required inflation factor based on 12 participating adolescents per team is calculated by:  1 + (m-1) ?= 1 + (12-1)0.02 = 1.22  An estimated (280)(1.22) or 342 adolescents will be required for each of the control group and intervention group. If there are 23 participating teams in each study group then an estimated 15 (NS) players per team will be required. 126  Appendix B Consent Form   127    128    129 Appendix C Preseason Questionnaire   130    131 Appendix D Physical Activity Readiness-Questionnaire   132  133 Appendix E 2008-2009 Microsoft Access Injury Survey      134  Appendix F Weekly Exposure Sheet  135  Appendix G Injury Report Form     136    137    138 Appendix H Therapist Assessment Form   139 Appendix I Physician Assessment Form   140 Appendix J Injury Prevention Program Exercise Card     141    142 Appendix K Soccer Injury Prevention Program Brochure  143 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0074175/manifest

Comment

Related Items