km...

‘il
"a

u.‘
‘ i r
“5?:

, I

 

rot! .
6 #4: a

an»? .

s I. I.

. r51.
var : 1
139%.?“

‘ _ .21.

 

.Hv.té.:;.a§. , . A . .Mu133. .flafi , ‘ %I.I.:.lv
‘ . . V4 , .

a a.fi§..,§.

LIBRARY
Michigan State
University

This is to certify that the

dissertation entitled

INJURY IN YOUTH FOOTBALL: PREVALENCE,
INCIDENCE, AND BIOLOGICAL RISK FACTORS

presented by

Peter J. Morano

has been accepted towards fulfillment
of the requirements for

PhoD. degree in KineSiOIOgy

flQVW/MWML

Major professor

Date 3 [374° 3

MS U is an Affirmative Action/Equal Opportunity Institution 042771

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

1'0 '18 ’6 5
APR 0 6 2r]

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c-JCIRC/DatoDuepSS-nts

 

 

INJURY IN YOUTH FOOTBALL: PREVALENCE, INCIDENCE, AND
BIOLOGICAL RISK FACTORS

By

Peter J. Morano

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Kinesiology

2003

ABSTRACT

INJURY IN YOUTH FOOTBALL: PREVALENCE, INCIDENCE, AND
BIOLOGICAL RISK FACTORS

By

Peter J. Morano

PURPOSE: To estimate injury rates and examine biological risk factors
(height, weight, body mass index, and maturity status) as predictors of injury
in a youth football population. Surveys of youth participants generally lack
suitable exposure data for practices and competitions to permit estimates of
injury rates. METHODS: Participants (n=354, 9-14 years) in two community
youth football leagues in mid-Michigan were followed throughout a single
season, mid-August through October. Grade in school was the unit of
competition. A certified athletic trainer was on sight for practices and home
games to record exposure statistics and injuries as they occurred; injuries
occurring in road games were reported by coaches and verified the next day
by the trainers. Height and weight were recorded at the beginning of the
season with which body mass index was calculated. Informed consent from
both parents and child were obtained from 296 participants. The informed
consent allowed use of the biological parents’ heights in order to predict the
adult height of the athlete to estimate maturity status. A Risk of Injury in Sport
Scale (RISSc) questionnaire was issued to and completed by the participants

in order to examine psychological variables and the likelihood of injury. The

RISSc questionnaire was composed of 24 questions subdivided into 6 factors
(uncontrollable, controllable, upper extremity, surface related, overuse, and
reinjury). Also examine was a correlation analysis between the RISSc results
and the biological variables. RESULTS: Estimated injury rates were
significantly higher in 7-8th (1602/1000 athlete exposures) compared to 4-6th
(7.5/1000 athlete exposures) grades (p<0.001). Injury rates per 1000 athlete
exposure (AE) were higher among the 7th (18.2/1000AE) and 8‘“
(17.6/1000AE) grade teams than the 4/5th (8.2/1 OOOAE) and 6th (9.6/1 OOOAE)
grade teams. Late maturing individuals in the younger (4"‘—6th grade)
population showed an increased likelihood of injury compared to on-time and
late maturing players (p <.05). Having a high body mass index (BMI) was
shown to be a predictor of injury in the younger group while the uncontrollable
and reinjury RISSc factors were shown to be predictors of injury in the older
group. CONCLUSION: Estimated injury rates increase with grade and
probably age. Biological variables and a player’s perception of risk of injury
seem to partially influence the likelihood of being injured. There was no
evidence of interactions of internal variables with the likelihood of injury in

youth football.

To my parents and daughter:
Pete Sr., Barbara, and Sophia
“If you don't know your family's history, than you don't know anything”.

--Michael Chrichton

iv

ACKNOWLEDGMENTS

Dr. Robert Malina, Tarleton State University. Thank you for your continued
guidance, vision and friendship —I am proud to be associated with you and your
work. You’ve really opened my eyes to new career possibilities and interests. I
value our relationship and I really enjoyed your social gatherings and getting to
know your family.

Dr. John Powell, Michigan State University. You were a great addition to MSU
and to me personally. Your knowledge and mentoring has helped me in
numerous ways. I appreciate your patience and everything you have done for
me.

Dr. Jeff Kovan & Dr. Larry Nassar, Michigan State University. You two rounded
out what, in my opinion, was the greatest graduate committee ever assembled in
the KIN department at MSU. Thank you for your help and friendship throughout
my academic career.

Susan Drabik & Emma Drabik-Marcus. You have done more for me than I could
ever give in return. Thank you for your love, friendship, understanding, and
dinners - I cherished the time we spent together©

Dr. Harold Marcus. It was your initial inspiration that led to my current career.
You were more than an uncle and I will always appreciate the advice and comfort
you gave.

Dr. Sean Cumming, University of Washington, Dr. Anthony Kontos, University of
New Orleans, & Mr. Tom Dompier, Michigan State University. You all helped me
immensely - personally and professionally. We had some great times as
graduate students together. I know our friendship will last and I look fonNard to
many years of personal and professional interaction.

Mr. Anthony Moreno, Mr. Kevin Stefanek, Michigan State University. Friends like
you truly made life as a graduate student as fun as possible.

Mr. Jeff Monroe, Mr. Tom Mackowiak, & Dr. Sally Nogle, Michigan State
University. It’s been 15 long years since I first came to you with interest in
athletic training. Thank you for your mentoring over the years. I hope to do the
program proud.

Dr. Ellen Whitford, Central Connecticut State University. Thank you for your
patience and understanding over the past 3 years. More importantly, thank you
for the opportunity to be a part of the Blue Devil family.

This research was funded in part by a grant from the National Athletic Trainers’
Association Foundation (Grant Number 300R001), Robert M. Malina (Principal
Investigator) and by a William Wohlgamuth Memorial Fellowship in the
Department of Kinesiology, Michigan State University.

And, to my family, Peter & Barbara Morano and Julie Allain. Your love and

support is immeasurable. This capstone is the result of your gift of education for
which I can’t thank you enough.

vi

TABLE OF CONTENTS

List of Tables ................................................................................. ix
List of Figures ............................................................................... xi
Chapter 1
Introduction .................................................................................. 1
Research Questions ................................................................ 6
Limitations .............................................................................. 8
Significance ............................................................................ 8
Chapter 2
Review of Literature ....................................................................... 10
Injuries in Youth Sports..................... ........................................ 10
Football Injuries. . . .' ................................................................... 21
High School Football Injuries ...................................................... 22
Youth Football Injuries ............................................................... 29
Age Grouping in Youth Sports .................................................... 39
Assessing Maturity Status ......................................................... 40
Summary of Literature Review ................................................... 45
Chapter 3
Methodology ................................................................................. 48
Population ............................................................................. 48
Project Design ........................................................................ 49
Data Collection ....................................................................... 51
Predicting Adult Height ............................................................. 54
Injury Rate Analysis ................................................................. 57
Statistical Analysis .................................................................. 58
Chapter 4
Results .......................................................................................... 62
Descriptives ........................................................................... 62
Analysis of Research Questions ................................................. 64
Chapter 5
Discussion .................................................................................... 71
Chapter 6
Summary and Conclusions ............................................................. 78

vii

Tables and Figures ........................................................................ 84

Appendices

Appendix I. UCRIHS Approval Letter ................................................... 104
Appendix II. Participant Consent Form ................................................ 105
Appendix III. Parental Informed Consent Form ....................................... 107
Appendix IV. Background in Sport Information ....................................... 109
Appendix V. Game / Practice Information Form ..................................... 111
Appendix VI. Injury Report Form ......................................................... 112
Appendix VII. Risk of Injury in Sports Scale (RISSc) ................................ 114
Bibliography .................................................................................. 117

viii

LIST OF TABLES

Table 1. Injury rates per 1,000 exposures by sex and level of competition of
numerous high school sports over an 8-year period .............................. 84

Table 2. Injury rates among high school football players by source of data and

definition of injury ........................................................................... 85
Table 3. Injury rates among youth football players by source of data and

definition of injury ........................................................................... 86
Table 4. Characteristics of players by grade and community ................... 87

Table 5. Mean corrected parental heights and midparent heights (cm) ...... 88
Table 6. First organized sports reported by the study population ............... 89

Table 7. Prevalence of injured players during the season by grade and age
group ........................................................................................... 90

Table 8. Types of reportable injuries sustained during the football season,
communities combined .................................................................... 91

Table 9. Frequency of reportable injuries by position or activity during the
season, communities combined ......................................................... 92

Table 10. Athlete exposures by practices, games, and both combined by
community ..................................................................................... 93

Table 11. Estimated practice rates per 1000 athlete exposures (AE) by grade and
community ..................................................................................... 94 .

Table 12. Estimated game rates per 1000 athlete exposures (AE) by grade and
community ..................................................................................... 94

Table 13. Estimated game and practice rates combined per 1000 athlete
exposures (AE) by grade and community .............................................. 95

Table 14. Estimated game and practice injury rates, and rates for games and
practices combined per 1000 athlete exposure (AE) by grade in the total
sample ........................................................................................... 95

Table 15. Estimated injury case rates and player rates per 100 players by
community, and communities combined ................................................ 96

ix

Table 16. Estimated player rates and case rates per 100 players by grade...96

Table 17. Odds ratios of being injured for upper (above the mean) and lower

(below the mean) values of height, weight and body mass index .................. 97
Table 18. Reliability of the RISSc questionnaire ....................................... 98
Table 19. Logistic Regression final model in 4th-6th grade ......................... 99
Table 20. Logistic Regression final model in 7th-8th grade ......................... 99

Table 21. Partial correlations for biological variables (height, weight, BMl,
maturity) and the Uncontrollable RISSc factor in the older population ............ 99

LIST OF FIGURES

Figure 1. Mean heights of youth football players relative to reference values for
United States boys ........................................................................... 100

Figure 2. Mean weights of youth football players relative to reference values for
United States boys ........................................................................... 101

Figure 3. Mean BMI’s of youth football players relative to reference values for
United States boys ........................................................................... 102

Figure 4. Mean percentage of predicted adult height in youth football players
compared to corresponding data for males from the F913 Research Institute
longitudinal growth study upon whom the height prediction equations were
based ............................................................................................. 103

xi

CHAPTER 1
INTRODUCTION

Organized youth sports programs annually enroll approximately 25 million
participants (Carnegie Corporation of New York, 1996). Participation has the
shape of a pyramid with youth at the broad base and the elite at the top; high
school and college athletes comprise the intermediate levels.

Risk of injury is inherent in sport and other activities of childhood and
adolescence, but systematic study of the incidence of injuries in organized youth
sports is still relatively limited. Even with the large numbers of youth involved in
organized sports, most of the past and current research on athletic injuries has
focused on the high school, collegiate, and elite levels of competition. Several
reasons for this focus on higher levels of competition could be that researchers
have easier access to the athletes, and/or that increased fan and financial
interests which make research with these populations more appealing and
fundable. It is also possible that injuries occur more frequently at more elite
levels of competition. Organizations which manage higher levels of competition
also have the financial resources to monitor injuries and perhaps fund research.

In addition to limited population-based research on youth sports, there is a
lack of information regarding body size, biological maturity status and behavioral
characteristics, alone or in combination, as potential risk factors for injury in
sports. Many studies of injury in sport concentrate on environmental variables or
external risk factors, such as field surface/conditions and equipment. This is

somewhat surprising since growth and maturity characteristics are often

indicated among player-related (internal) risk factors for injury in sports (Caine
and Lindner, 1990). External variables are easier to observe, control and/or
change if they are found to be unsafe.

The Consumer Product Safety Commission’s National Electronic Injury
Surveillance System (NEISS) regularly maintain sport-related injuries, but data
are limited to those who present to hospital emergency rooms for treatment
(NEISS, 1981). NEISS published data on product related (external variables
such as equipment) injuries for the 12-month period ending December 1981 and
indicated that for every 100,000 participants 5-14 years of age, there were 359
injuries in baseball, 280 in basketball, 454 in football, 101 in gymnastics, 108.1 in
soccer, 102.3 in wrestling, and 906.7 in bicycling. Player related variables such
as body size and composition, biological maturity status, physique, and so on
were not reported. Moreover, these player characteristics cannot be altered.

Athletes are sometimes grouped by internal or player-related variables,
but this depends on league rules. There has long been an assumption in the
medical community that grouping young athletes by size or maturity status may
help to reduce the number of injuries (Roser and Clawson, 1970; Kreipe and
Gewanter, 1985; Backous et al.; 1990, Malina and Beunen, 1996). Yet, these
issues have not received systematic study. Some authors even suggest that
contact sports for the physically immature are inherently unsafe, and do not
recommend participation for children (Godshall, 1975; Roser and Clawson,

1970).

Nevertheless, all indications show an increase in sports participation at
every age level. Since injuries occur in sports, the possibility of permanent
damage as a result of athletic injury is a legitimate concern. Previous
epidemiologic research has shown that good injury data can identify problem
areas in which preventive measures can be implemented to reduce the incidence
of injuries (Mueller and Blyth, 1982).

With regard to available studies of injury in sport, methodological concerns
about data collection and athlete exposure are of critical importance (Backx et
al., 1989; Watkins, 1996; Thompson et al., 1987). Some studies use
retrospective accounts of injury. Athletes and/or parents and sometimes
coaches are asked to recall an injury episode that occurred a day or sometimes
months previously. With time, the recall of events may not be accurate so that
details about severity, time loss, type of injury, and other factors (e.g., context)
may be compromised. Other retrospective studies use hospital or insurance
records, which may allow a detailed description of an injury, but such records
include only injuries that were severe enough to be seen by a physician or
reported to an insurance company. Injuries that are not seen by a physician, or
that are seen by a family physician, may be missed or not reported in this type of
data collection process. Moreover, personal experience indicates that many
injuries require a participant to lose playing time, but are not deemed serious
enough to report to a physician or hospital. Further, injuries associated with
sports equipment, but not in the context of a sport event or practice, are often

labeled as sport-related injuries in hospital and insurance records.

Studies that ask for weekly, monthly, or season ending reports from
coaches have potential problems with accuracy and reliability. Coaches differ in
detail of injury recording and may omit some injuries because of fear that their
team will be viewed as unsafe. Coach accounts of an injury episode may also
differ from what an athlete recalls. Given time constraints on coaches, many
simply do not complete or superficially complete such forms.

Lacking in each of these situations is a health care professional to record
injury data on-site on a day-to-day basis. Having a sports medicine professional
on-site provides more opportunity for more accurate and detailed injury
information.

Studies of injuries in a given sport vary with length of season and amount
of athlete exposure. An exposure is an opportunity for an athlete to sustain an
injury. Some studies define exposure by minutes, hours or sessions. The best
way to capture the true risk of injury in any sport is to accurately cover the sport
for at least an entire season, from preseason conditioning to competition. This
includes all practice and game sessions. Some studies of youth sports only
cover exposures, or opportunities for injury, during a tournament, especially
youth soccer. Tournament settings are atypical of the regular season.
Comparisons of injury data from a tournament with corresponding injury data
throughout an entire season have limitations, but allow access to a large group of
athletes in a very short period of time. Teams often play daily and sometimes
more than once a day for several days or an entire week. And, within

tournament settings, health care providers (commonly first aid technicians) are

usually available to the athletes in case of injury. A problem with this type of
service is quality control, internal consistency among health care staff, which may
influence the accuracy of the data.

A major methodological concern in sports injury surveillance (injury
reporting) is the definition of a reportable injury. There is no universally accepted
definition of an athletic injury (Thompson et al., 1987). Without consistency in
defining a reportable injury, comparisons of studies are difficult. The type of
study may determine the definition of a reportable injury. At other times,
researchers create their own definition to accommodate local needs. It is likely
that one definition of a reportable injury is not appropriate for all types of sports
injury studies mainly due to staff differences.

There are many potential risk factors for injury, whether or not a child or
adolescent actually sustains an injury. Risk factors are generally categorized as:
player/athlete-related and sport/environment-related risk factors. Player-related
risk factors include physique, structural alignment, flexibility, strength, motor skill,
proprioceptive / kinesthetic awareness, behavior, history of injury, the adolescent
growth spurt, maturity status and perhaps others with few exceptions. Player-
related risk factors are difficult to alter. However, athletes can be grouped with
other athletes with similar characteristics. The specific role of individual player-
related risk factors in injuries in youth sports is not well known (Malina, 2001 ).

Sport-related risk factors include training programs, playing conditions,
equipment, age groups (size, maturity, and experience mismatches in broad age

groups), coaching, parent behaviors, sport organizations, sport/league rules and

inadequate rehabilitation from a previous injury. Sport or environment-related
risk factors can be generally controlled. Even if an ideal sport environment were
created, would injuries be eliminated? Smith et al. (1993) suggest that one-half
of all injuries sustained in youth sports can be prevented. Unfortunately,
systemic evidence to document this suggestion is not provided by the authors.

An athlete’s perception of risk of the injury may also have an influence on
the likelihood of injury in sport. Perception of risk may be related to or
influenced by biological factors such as body size and skill level. Players who
are larger may feel less likely to become injured, especially in collision sports,
where size is usually considered an advantage. Therefore, this child may
partake in behaviors involving more risk on the field. Also, an athlete who
possesses high efficacy about his skill level may also take more risks during
practices or games. Athletes with high self-efficacy may view injury as a
negative episode and think that the chances of injury are unlikely (Kontos, 2000).
Research Questions

The present study concerned injury in a sample of sports participants 9-14
years of age in a single sport — American youth football - over the course of a
single season. Several specific questions were the focus of the study:

1. What are the prevalence and incidence of injury in youth football?

2. Do injury rates differ by grade?

3. Are there biological variables (weight, height, biological maturity) that
are associated with the risk of being injured, and are there interactions among

these variables?

4. Is the perception of risk of injury in youth football related to the
likelihood or frequency of injury?

The following hypotheses were tested:

1. Injury rates increase with grade.

2. Weight, height, the body mass index, and biological maturity by
themselves do not increase the likelihood of injury in youth football.

3. Player’s perception of risk influences the likelihood of injury.

4. Internal variables (player-related factors) interact with each other to
influence the risk of injury in youth football.

Questions can be raised and variables can be examined in a child and
adolescent population that cannot be considered with older subjects. Athletes 9-
14 years of age show considerable variation in size and maturity status more so
than during childhood and later adolescence. It is within this age range that boys
enter adolescence, with associated changes in size, physique, body composition,
performance, and behavior. Many boys will experience their maximum rate of
growth, reaching peak height velocity (PHV), during this interval (Malina and
Bouchard, 1991). Changes in size and physique, and individual differences in
maturational timing are potential risk factors that are impossible to assess in
older adolescent populations because most of the subjects will have already
passed through the spurt and are close to physical maturity.

A unique feature of the present study is that it examines the risk of injury
in youth football at the community level, a level that is not commonly considered

in sport injury research. In addition to exposure data for practices and games for

participants in two communities throughout a season, age, height, weight, the
body mass index (BMI), estimated biological maturity status, and perceived
likelihood of becoming injured in youth football are considered.
Limitations

A prospective, day-to-day analysis of athletic injuries in a youth football
population is a new experience for athletes and coaches, as well as for certified
athletic trainers (ATC), the primary data collectors. Certified athletic trainers are
allied health professionals whose main job is to prevent, evaluate, treat, and
rehabilitate athletic injuries. Most coaches and athletes at the youth level have
never had the service of a certified athletic trainer and are unfamiliar with the role
an athletic trainer plays with a sports team. This lack of knowledge about athletic
trainers can contribute to inconsistency in data collection. Some athletes may
not report an injury to the ATC for several days to weeks after initial onset.
Coaches might not report injuries to the ATC. Parents may have uncertainty
about the purpose of an ATC and may not cooperate with the trainer’s evaluation
of their child’s injury.
Significance

The significance and implications of the study takes several directions.
First, it provides season-long data for a single sport at the local level, including
practices and games. Second, all data were recorded by two certified athletic
trainers. And third, the study used a non-invasive method of estimating biological

maturity status.

The results of the study can potentially be useful to parents, coaches, league
organizers, and athletes involved in youth football, and perhaps to other sports.
There may be a better understanding of the risks involved in youth football. The
league, coaches, and parents could realize the benefit of having a certified
athletic trainer on site to handle injuries on a day to day basis which could
include a decreased loss of playing time, fast assessment of injuries, and less
hassle in dealing with injuries themselves. If a non-invasive method of
assessing physical maturity proves to be useful as well as being predictive of the
likelihood of injury for youth football, it could be incorporated into youth leagues

that have such concerns.

CHAPTER 2

REVIEW OF LITERATURE

Several types of literature need to be reviewed in order to understand the
complexities of injury research with youth. Certainly, the literature on sports
injuries, especially injuries in American football, is relevant. Because the
population in this study is of child and adolescent age, specific sports injury
studies of the same age and of the same sport may be limited in availability. An
extensive literature search revealed that most sports injury surveillance studies
deal with high school or college athletes; therefore, injury data on youth football
is limited. With this in mind, any literature on child and adolescent sports injuries
can be of value. Since biological maturity status is a potential risk factor for
injury, the literature on growth and maturation are also of importance.

Injuries In Youth Sports

In general, injury data on youth sports are collected retrospectively using
general survey design. The Child Health Supplement to the 1988 National
Health Interview Survey (NHIS) conducted by the United States National Center
for Health Statistics, for example, provided estimates of the incidence of injuries
associated with sport and recreational activities in 1988 (Bijur et al., 1995). The
survey considered injuries to children and adolescents 5-17 years of age who
received medical attention during the year surveyed. An adult in the surveyed
household reported the data. Injuries were defined as “...those that occurred in a

place of recreation and sports" (Bijur et al., 1995, p. 1010). The overall incidence

10

of injury from sports and recreation was 6.4/100 players. There was a sex
difference in the estimated incidence, 8.3/100 players for males and 44/100
players for females. The injury rate was stable in boys from 5 to11 years,
increased threefold from 11 to 15 years, and then declined after 15 years of age.
Injury rates for girls were steady from 5 to 8 years of age, increased fourfold from
8 to 12 years of age, and then declined from 13 to 17 years. The most
commonly reported injuries were sprains (1 .9/100 injuries), fractures and
dislocations (1 .7/100 injuries), and contusions (0.6/100 injuries). When the data
were grouped into age categories (5-9, 10-13, and 14-17 years), fractures were
consistent at around 25% of the total sports injuries reported. The percentage of
lacerations and sprains changed positions from the youngest age group to the
oldest. In the 5-9 year age group, lacerations accounted for 43% of the total
injuries and sprains accounted for 10%. Among youth 14 to 17 years, sprains
accounted for 30% while lacerations accounted for 22% of all injuries (Bijur et al.,
1995). The NHIS survey did not include information that would allow estimates
for injuries in organized sports and specific sports, for exposures, and for sport
specific contexts of injury.

The Canadian Hospital Injury Reporting and Prevention Program
(CHIRPP) is an emergency room-based injury surveillance program constituting
ten pediatric hospitals. The CHIRPP definition of an injury was "...any injury
incurred while the victim (age 5-19) was engaged in a physical activity for which
the main purpose was competition, practicing for competition or improved

physical health; the competition or practice could have been formal or informal”

11

(Ellison and Mackenzie, 1993, p. 96). Sport injuries so defined accounted for
25% of total injuries in the CHIRRP database. Consistent with United States
interview data, sports injuries presented to emergency rooms increased with age,
reached a peak at 11-13 years in girls and 13-14 years in boys, and occurred
more often in boys than girls by a ratio of more than 2:1. The three most
common sports injuries were (1) sprains/strains (32%), (2) fractures (21%) and
(3) hematoma (19%) (Ellison and Mackenzie, 1993).

Kvist et al. (1989) surveyed sports injuries in children and adolescents 6-
15 years treated in the Turku (Finland) University Central Hospital over three
years from 1980 to 1982. Sports injuries accounted for 21% of all injuries in this
age group with nearly one-half occurring during the winter months. This is most
likely due to a long Finish winter. Boys were injured more than girls by a ratio >
2:1. Sport injuries increased with age, and reached a peak at 12-13 years in girls
and 14-15 years in boys. The three most common injuries were (1) fractures
(26%), (2) sprains/strains (24%) and (3) contusions (22%) (Kvist et al., 1989).

Consistent in all three surveys were the three most common injuries —
contusions, sprains/strains, and fractures. Percentages differed slightly among
studies, but the data suggested that geography and sport did not differ greatly in
the type of injury. Another consistency between the CHIRPP and the Finnish
studies was that contact of some kind during play or sport was involved in most
of the injuries, whereas contact played lesser role in the NHIS. The NHIS study,
however, included injuries due to bicycles, animals, other vehicles, playgrounds,

struck/fall, overexertion, and other recreational activities (Bijur et al., 1995).

12

lntuitively, it can be assumed that contact was involved in bicycle, other vehicle,
animal, and playground accidents, thus increasing the number of injuries that
occurred as the result of some kind of contact. Quite often, injuries occur from
bicycle and automobile collisions with other like vehicles, trees, poles, and so on;
falling off playground equipment; and contact with aggressive animals.

An often-cited study of sport injuries in youth is that of Zaricznyj et al.
(1980). Sport related-injuries in school age children in a midwestern community
were followed for one year. The population consisted of 25,512 school children
in grades K-12 (aged 5 years and older). Comparisons were made between
elementary, junior high, and senior high school students, and among non-
organized sports, physical educations classes, community team sports, and
school team sports. Sources of data were the aggregate of (1) hospital records
from two local hospitals, school accident insurance forms and local physicians
reports, and (2) reports of principals, coaches and supervisors of community
sports programs. An injury was defined as “...any traumatic act against the body
sufficiently serious to have required first aid, filing of school insurance accident
reports, or medical attention” (Zaricznyj et al., 1980, p. 318). A standard time
loss definition of injury was not used because non-organized sports were
included in this study.

Of the 25,512 school children comprising the study population, 1,495 (6%)
sustained 1,576 injuries during the one-year period. Twenty-two percent (353) of
the injuries occurred in elementary school children (5-11 years), 18% (282)

occurred in junior high school students (12—13 years), and 59% (930) occurred in

13

senior high school students (age 14+ years). Comparison by school population
showed that sport-related injuries occurred in 3% of the total elementary
population, 7% of junior high students, and 11% of senior high school students.
The highest incidence of injury was at 15 years in boys (15%) and 14 years in
girls (8%), but boys sustained twice as many injuries as girls when the data were
combined across grades (Zaricznyj et al., 1980).

Non-organized sport activities produced the most injuries (40%), followed
by physical education classes (38%). Organized school sports accounted for
15% of the injuries while community team sports accounted for the remaining
7%. Among organized sports, football accounted for four times more injuries
than any other sport with 148 injuries (126 in school sanctioned football and 22
from community based football). Football also had the most injuries when levels
of play were combined (295 injuries, 9.9% of all injuries).

Tursz and Crost (1986) examined sports-related injuries in children in a
community in France. Data were collected on all accidents involving children
aged 0-15 years (upper age limit in the French pediatric wards). An accident was
defined as “...an unexpected, unintentional, and violent event affecting a child,
with or without detectable lesion, and subsequently leading to medical
attention...” (Tursz and Crost, 1986, p. 294). Data were mainly collected from
two general public hospitals, their two mobile emergency and resuscitation units,
and eleven private hospitals. Data were not recorded on level of competition or

intensity of training. Very few cases of sport-related injuries occurred in children

14

 

under 6 years of age; therefore, the results reflected only injuries sustained to
children between of 6 and 15 years (Tursz and Crost, 1986).

There were a total of 7,182 accident cases recorded in the two public
hospitals, the two mobile emergency and resuscitation units, and four of the 11
private hospitals that were most important in size and number of patients.
Overall, 789 sports accidents were recorded, 597 (76%) in out-of-school sport
activities and 192 (24%) in school physical education. Boys sustained more
injuries than girls by a ratio > 1.5:1. Boys accounted for 487 sport accidents
(62%) compared to 302 (38%) for girls. Out-of—school sporting activities resulted
in more accidents in boys than girls, but there was no difference in school sport
accidents. Fifty-three percent (421) of all sports injuries occurred to children 12-
15 years of age. After the age of 6 years, sports accidents accounted for 17% of
all reported accidents (Tursz and Crost, 1986).

Contusions and fractures in girls, and contusions and cuts/lacerations in
boys were the accidents most frequently reported. In both sexes, contusions and
sprains/strains increased with age, while cuts/lacerations decreased with age. A
fall was the reported cause of injury in 58% of the sport - related accidents,
followed by being struck by an object 16% and by being struck by another child,
9% (Tursz and Crost, 1986). These contexts can be categorized as contact,
resulting in 78% of sports-related injuries in this survey.

Sport injuries in school-aged children were examined over a six - week
period in the Netherlands in 1982 (Backx et al., 1989). Two classes from 175

schools were randomly selected. A total of 7,707 students, 8-17 years,

15

comprised the study population. An injury was defined as “...physical damage
caused by a sports-related incident and reported as such by the respondent...”
(Backx et al., 1989, p. 235). A questionnaire was distributed to the students
during physical education classes and completed with assistance of the teachers.
A total of 7,468 students completed the questionnaire (96.9% of the study
population). The first part of the questionnaire requested background information
(age, sex, extent of sporting activities, and so on). The last question of the first
section asked about newly sustained sport injuries. The second section
contained detailed questions regarding these injuries, and was completed with
the help of a teacher or a parent (Backx et al., 1989).

A total of 732 students accounted for 791 total injuries in the study
population over a six - week period. A diagnosis was given in about 50% of the
sport - related injuries. Of those with a diagnosis, 40% were sprains, 37% were
strains, 7% were fractures, and 2% were concussions. Other types of injuries
accounted for 8%. Most of the injuries were sustained during club sport
activities: 29% in training sessions and 33% during matches. Physical
educations classes accounted for 21% of sport-related injuries while non-
organized sports accounted for 17%. The total incidence rate was 10.6/100
participants over the six - week period. Boys had a greater risk of injury than
girls, 1.38:1. There also was a positive correlation with injury risk and age.
Students 15 and 16 years had the highest risk ratio, 2.09, compared to students

8 to 10 years (Backx et al., 1989).

16

Basketball and field hockey had the highest risk ratio for both boys and
girls in the Dutch study (1.99 and 1.83, respectively). Girls were twice as likely to
become injured in field hockey than boys (Backx et al., 1989). Age categories
were not determined for each individual sport. Unlike Zaricznyj et al. (1980),
Backx et al (1989) found more injuries in organized sports than in non-organized
sports and physical educations classes. Organized sports are more likely to
have better recording methods. Note, however, the data were based on
questionnaires completed by the students, which may have limitations. Like
previous surveys on sport injuries to youth, only injuries serious enough to be
reported were included.

Watkins and Peabody (1996) retrospectively examined sports injuries in
children and adolescents who were treated at a sports injury clinic in London,
England. Data were collected on all patients 5-17 years from January 1, through
December 31, 1989. The clinic treated 394 sports injuries during the year. They
represented 14% of all injuries seen at the clinic. Of the 394 sports injuries, 89%
(351) occurred in training or competition in organized sports and 11% (43)
occurred in free-play or recreational activities. Males sustained 55% (216) of the
sport injuries compared to 45% (178) in females (Watkins and Peabody, 1996).
The frequencies of injuries increased with age and peaked at 13-14 years in girls
and 15-16 years in boys. Other studies (Bijur et al., 1995; Ellison and
Mackenzie; 1993, Zaricznyj et al., 1980) also reported that boys sustain more
injuries than girls at a rate of almost 2:1. The ages at highest frequency of injury

in this study are consistent with previous surveys.

17

Soccer players were the most frequent male athletes treated at the
London clinic, followed by rugby players. This, of course, reflects the popularity
of these sports in England. Male soccer players sustained 48 injuries (22% of all
sports injuries) and rugby players accounted for 36 injuries (17%). The most
frequently treated female athletes at the clinic were those participating in athletics
(36 injuries, 20%) and gymnastics (23 injuries, 13%). The authors attributed the
sport - specific frequencies to two factors: a) the risk associated with the
particular sports, and (b) the number of participants involved in the particular
sports (Watkins and Peabody, 1996). It is no surprise that soccer and rugby
accounted for most of the injuries in males, and athletics and gymnastics
accounted for most injuries in females because they were the two most popular
sports for each gender in the United Kingdom (Watkins and Peabody, 1996).

A survey of athletic injuries treated at an athletic medicine facility in
upstate New York was reported by DeHaven and Lintner (1986). The study
population was drawn from students at the University of Rochester (including
athletes on NCAA Division III sport teams) and from the surrounding community
over a period of 7 years from June 1976 through June 1983. The sample was
categorized by age, sex, and sport of injury. The database consisted of 4,551
total cases with information on age, gender and sport for 3,431 of the cases
(DeHaven and Lintner, 1986). Ages of the subjects ranged from < 13 years to 80
years. The authors did not specify the range for those under 13 years and
treated these as a single age group. The number of injuries treated at the clinic

increased from < 13 years to 16-19 years and then steadily decreased in every

18

subsequent age category except for 31-40 years, which had more injuries than
the 25-30 year age group. The 16-19 year age group had the most injuries,
1,408, more than double the number among 13-15 year olds, 514 injuries. Age
categories were not further subdivided by sex. Of the 3,431 cases treated at the
clinic, males accounted for 80% (2754) and females 20% (677) of the injuries
(DeHaven and Lintner, 1986).

The sports in which the injury occurred were also documented. Football
accounted for 2,193 injuries treated at the clinic, more than 12 times the number
observed in the next most common sport, basketball (DeHaven and Lintner,
1996). This can be attributed to the larger number of athletes in football and the
fact that the intercollegiate athletes at the University of Rochester were treated at
the clinic.

Beachy et al. (1997) examined male and female high school students in
grades 7 through 12 at a private school in Honolulu, Hawaii. A total of 14,318
athletes in 32 sports participated in the study over an 8-year span. Multi-sport
athletes were counted once for each sport. An injury was defined as "...any
athlete complaint that required the attention of the athletic trainer, regardless of
the time lost from activity. Five injury classifications were used: 1) minor, no time
lost; 2) mild, 1-7 days lost; 3) moderate, 8-21 days lost; 4) severe, 22 or more
days lost; 5) catastrophic, permanent disability, dismemberment, or death...”
(Beachy et al., 1997, p. 676). Sources of the injury data were evaluations by two
full-time athletic trainers, one part-time athletic trainer, and two team physicians,

all of whom were constant throughout the study.

19

 

Over the 8 years, there were a total of 11,184 injuries (7,026 in males and
4,158 in females). The athletes were divided into four levels of competition:
intermediate (INT) - grades 7-9, junior varsity (JV) — grades 9-11, varsity/junior
varsity (V/JV) — grades 9-12, and varsity (VAR) - grades 9-12. Table 1 shows
injury exposure rates per 1000 exposures of each level of competition by sex.
Estimated rates of injury were about equal for boys and girls. The sport with the
greatest number of injuries and highest risk of injury was football (2,503 and
0.80, respectively). For girls, track had the highest number of total injuries at
1120, but soccer had the highest injury rate, 0.71/1000 exposures (Beachy et al.,
1997)

The length of the study and number of subjects provided good data. The
stability of the medical staff over the entire 8 years added to the reliability and
validity of injury reporting. However, the definition of a reportable injury may not
reflect the true rate of injuries sustained by this population. Any time an athlete
had a health complaint and sought the attention of an athletic trainer, the episode
was considered an injury regardless of the severity of the complaint. The
definition of injury in this study was designed to emphasize the daily workload of
a certified athletic trainer in the high school setting, which may have inflated the
number of reportable injuries. A more precise label for the term injury, used in
this study, would be athlete-trainer contact, which better represents the workload
of athletic trainers in the high school setting.

Kontos (2000) was apparently the first to examine the combination of

perception of risk, risk taking behaviors, and body size in the context of youth

20

sport injuries. Two hundred and fifty two (142 male, 111 female) youth soccer
players 11 to 15 years of age were surveyed regarding their perception of risk of
injury in sport, risk taking behavior, and perceived skill level. Injury data were
collected via phone conversations with coaches twice per week over an eight-
week period.

Results suggested that the BMI, in part, influenced levels of perceived
risk. Subjects with a higher BMI showed a higher level of perceived risk than
those with a lower BMI. In soccer, being heavy is not an advantage. There is a
general negative stigma in American society associated with a child being heavy
and this may lead to heavy youth soccer players to view themselves as being
less skilled, more awkward, and more likely to be injured playing soccer. The
BMI was the best predictor of injury in this study. Soccer players with a high BMI
(upper tertile of the study population) significantly sustained more injuries than
players with a lower BMI in the lower tertile (Kontos, 2000).

In contrast to soccer, heaviness is often perceived to be an advantage in
American football. Youth football players may exhibit a different perception of
risk of injury than soccer players simply because of the nature of the sport.
Football Injuries

Football is classified as a high-risk sport, which implies that the sport, by
its very nature, will produce a high number of injuries (Petersmarck, 1998;
Culpepper and Neimann, 1983). Other sports that also fall under this umbrella
are ice hockey and wrestling. These sports share high amounts of contact (body

to body contact and body to playing surface contact). Football generally receives

21

the most publicity with regard to injuries because of the popularity of the sport

and the high numbers of participants from the youth through professional levels,

and the rare catastrophic injury.

High School Football Injuries

A three - year injury surveillance study conducted by the National Athletic

Trainers Association (NATA) examined 10 high school varsity sports, including

football (Powell and Barber-Foss, 1999). Certified athletic trainers collected the

data prospectively from 1995-1997. Data collection included reportable injuries,

athlete exposures, time loss due to injury, injury location, and injury type. A

reportable injury was defined as follows:

“Any injury that causes cessation of participation in the current game or
practice and prevents the player’s return to that session

“Any injury that causes cessation of a player’s customary participation on
the day following the day of onset.

“Any fracture that occurs, even though the athlete does not miss any
regularly scheduled session

“Any dental injury, including fillings, quations, and fractures.

“Any mild brain injury that requires cessation of a player’s participation for
observation before returning, whether in the current session or the next

session” (Powell and Barber-Foss, 1999, p. 278).

An average of 133 high schools per year reported football injury data. Over the

course of the study, 400 team-seasons, 21,122 player seasons, and 1,300,446

athlete exposures were reported. An athlete exposure occurred any time and

22

athlete participated in a practice or game session, and had an opportunity to
sustain an injury, regardless of the actual amount of time the athlete participated
(Powell and Barber-Foss, 1999).

Football had the highest injury rate of the 10 sports surveyed (baseball,
boys and girls basketball, boys and girls soccer, wrestling, field hockey, softball,
and volleyball). Case rates (number of injuries divided by the study population),
player rates (number of injured players divided by the study population) and
exposure rates were calculated. Case rates included multiple injuries where
player rates do not. Exposure rates were the total number of injuries divided by
exposures (total exposures, practice exposures, and game exposures).

Injury rates for high school varsity football were as follows: case rates -
50.0/100 players and player rates - 34.6/100 players. Total exposure rates were
8.1/1000 athlete exposures. Even though practices accounted for over 56.4% of
the injuries to high school football players, game rates, estimated at 26.4/1000
athlete exposures, were five times higher than estimated practice rates,
estimated at 5.3/1000 athlete exposures. Most injuries (72.5%) occurring to high
school football players were categorized as minor (loss of 1-7 days). Moderate
injuries (loss of 8-21 days) accounted for 16.3% and major injuries (> 22 days
lost) accounted for 11.2%. The three most common types of injuries were
sprains (31 .7%), general trauma (25.2%), and strains (21%) (Powell and Barber-
Foss,1999)

An early survey of football injuries, the North Carolina High School

Football Injury Study (Blyth and Mueller, 1974), examined injuries in 43 high

23

schools with 8,776 participants from 1969 through 1972. Trained investigators
who visited the schools on the same day and time each week of the football
season collected the data. Players who were injured the week before the
investigators made a visit were interviewed to ascertain details of the injury. The
definition of an injury was when medical treatment was sought or usual activity
was restricted for at least one day beyond the injury (Blyth and Mueller, 1974).

The incidence over the combined four years of the study was 48.8%.
Injury rates for each year did not differ greatly, with 1969 having the highest
(53.2%) and 1970 having the lowest (47.5%) rates. Injuries were distributed
about evenly between games and practices. Estimated incidence rates
increased with age, height, and weight, but interrelationships among these three
variables were not considered (Blyth and Mueller, 1974).

A study of high school football players in Texas (n = 4,399 from 75
schools) revealed that almost one-half of the athletes competing at the varsity
level sustained an injury during the 1989 football season. The sources of injury
data were reports form certified athletic trainers who worked for the high schools.
The athletic trainers were present at each practice and game, and served a
primary role in the valuation and treatment of the injuries. A reportable injury was
defined as one that “...occurred in football and meeting one or more criteria: 1)
any injury that causes a student athlete to miss all or part of a single practice or
game; 2) any injury (including dental) that is treated by a physician; and 3) all

head injuries reported to the athletic trainer...” (DeLee and Farney, 1992, p. 576).

24

This definition, according to DeLee and Farney (1992), was deemed as
more comprehensive than a time-loss from participation or operational definition.
DeLee and Farney (1992) gave two examples where the definition would include
injuries that a time—loss definition would not. If a player sustained a concussion
at the end of a Friday night game, the symptoms may have been resolved by
practice on Monday. Second, if a player fractured a finger (offensive tackle), he
may not miss any playing time due to the position played, whereas another
athlete with the same injury (quarterback) would miss playing time (DeLee and
Farney, 1992).

Over the course of the season, there were a total of 2,228 injuries (1242 in
games, 986 in practices). This yielded an estimated incidence of 0.51 injuries
per athlete per year. The estimated incidence of reportable injuries, as defined,
was 0.003 injuries per hour of exposure (3/1000 hrs) per athlete.

DeLee and Farney (1992) did a betterjob describing injury type and
anatomical location than many other studies. The source of the data was more
consistent than other studies, where health professionals with the same
credentials collected the injury data. One limitation was that the age of the
subjects was not defined and incorporated into the analysis. Only the term
varsity athlete was used, and this could include boys between 14 and 18 or 19
years of age.

A comparison of injury rates of high school football players playing on
natural and synthetic surfaces was conducted by Adkison et al. (1974). The

population included players participating on varsity football teams in 73 high

25

 

schools in Seattle and Spokane, Washington, and Portland, Oregon, during the
1971 season. The total number of players was not given. A coach or student
athletic trainer supervised by the coach, recorded injury and player data. An
injury was defined as “...traumatic medical condition sustained during game play
resulting in any one (or more) of the following: (1) necessitated discontinuation of
participation for the remainder of the game (2) resulted in missing two or more
practices (3) resulted in missing one or more subsequent games” (Adkison et al.,
1974, p. 132). Injuries sustained during practice sessions were not counted.

A total of 349 injuries were sustained in 660 games during the 1971
season. Games played on natural grass surfaces (n = 424) accounted for 218
injuries with an estimated rate of 0.51 injuries per game on natural grass. There
were 131 injuries to players competing on synthetic surfaces (183 games) with
an estimated rate of 0.63 injuries per game. Total injury rate was 0.53 injuries
per game (Adkison et al., 1974).

Adkison et al. (1974) did not include injuries sustained during practice
sessions and, therefore, were not able to calculate injury rates for an entire
season. Also, the actual numbers of participants was not included thereby
making it impossible to calculate player rates or case rates. Data were lacking
for number of participants and injuries occurring during practices, which would
have been easy to amass allowing for a more detailed analysis of injury statistics.

Culpepper and Neimann (1983) conducted a study of high school football
injuries in Birmingham, Alabama. The purpose of the study was to develop

profiles of football injuries on high school athletes. Data were obtained from

26

athletes treated at the Sports Medicine Clinic at the University of Alabama in
Birmingham from 1976 through 1979. An injury was defined as “...any traumatic
change in a student athlete’s medical condition sustained during a regular
practice session or game and for which professional treatment was received at
the Sports Medicine Clinic...” (Culpepper and Neimann, 1983 p. 873).

A total of 1,877 injuries were treated at the clinic during four football
seasons. The types of injuries sustained most often were sprains (32.2%),
contusions (24.8%), and strains (12.4%). Running backs were the most
commonly injured (19.6% of the injuries treated at the clinic), followed by
offensive tackles (11.6% of the injuries). The body parts sustaining the most
injuries were the knee (22.2%) and shoulder (13.3%). More than 17% of the
injuries were to players <15 years of age. Injuries to other age groups were not
counted (Culpepper and Neimann, 1983).

Violette (1976) was one of the few to examine the relationship between
biological maturity and injury in football. The data were from the first year (1969)
of the North Carolina study discussed previously (Blyth and Mueller, 1974).
Subjects included junior high school (13-15 years) and high school (15—19 years)
football players from 43 schools in North Carolina. The criterion for maturity was
a composite score of three secondary sex characteristics: testicular volume,
pubic hair, and axillary hair. Testicular volume and pubic hair were assessed on
a 6-stage scale and axillary hair on a 4-stage scale. The maturity index

combined the three assessments: 6+6+4 divided by 3. Along with the maturity

27

index, height, weight, the subscapular skinfold, and grip strength were measured.
The definition of injury was the same used for the North Carolina study.

Injury prevalence for junior high school football was estimated at 0.40 for
13 and 14 year old and 0.14 for 15 year old players. It was assumed that 15-
year-old players had increased exposure to injury due to their experience; yet the
injury prevalence was lower. It was concluded that 15-year-old junior high school
football players were probably the cause of injuries to younger, less mature
players. In contrast, 15-year—old players at the high school level had an
estimated prevalence of 0.34 injuries. This group, however, was less biologically
mature than the older players and was less exposed to the chance of injury
because of their inexperience at the high school level (Violette, 1976).

Injured football players at the junior high school level were slightly less
biologically mature than non-injured players within each age group. The
maximum possible maturity index was 5.33. No one participating at the junior
high school level had a maturity index above 5.00. The index ranged from 1.33-
4.67 at 13 years, from 2.00-5.00 at 14 years, and from 2.00-5.00 at 15 years.
The mean maturity index increased with age at every level (Violette, 1976).

Junior high football players at each age who measured low in subscapular
skinfold, height, and weight sustained more injuries than athletes who had larger
measurements. Increased grip strength was associated with increased injury in
maturity the 14-year-old age group. On the other hand, 13 and 15 year old
players who measured low on grip strength sustained more injuries than those

who were stronger (Violette, 1976). These findings indicate that a smaller body

28

size is associated with an increase chance of being injured. However, there
were no clear trends regarding risk of injury and strength. Interrelationships
among the variables and risk of injury were not analyzed.
YOUTH FOOTBALL INJURIES

According to Godshall (1975), there has been much criticism regarding
safety and risk of injury in youth football. Godshall states that both lay people
and physicians may be misleading in their reports because the conclusions are
not based on objective data, but on general impressions and unfounded
assumptions, and are perhaps influenced by the media on high school, college
and professional athletes.

There are very little objective data on injuries sustained in youth football.
After a review of mission, rules, and injury data made available from the Pop
Warner Football League, Godshall (1975) concluded that the benefits of youth
football far outweighed associated risks, not just the risk of injury, but also
emotional stress, problems with parents, and interference with junior high school
athletics. Roser and Clawson (1970) similarly stated that there is opposition to
youth participating in contact sports from individuals (lay people and health care
providers) who say that football poses too great a risk of injury. However, there
are little reported data to validate these claims.

Godshall expressed these concerns in 1975. Yet, 27 years later there has
been relatively little progress in gathering objective data on injury rates and risk
of injury in youth football. The most recent report dealing exclusively with youth

football injuries was that of Goldberg et al. (1988), who conducted a prospective

29

study of six New England Pop Warner Football Leagues. The study included
5,128 boys 8-15 years of age from 257 teams. The boys were grouped by age,
weight, and skill level into five divisions:

Junior Pee Wee - 8-11 years, 22.5 — 38.8 kg (n=692),

Pee Wee - 9-12 years, 29.3 - 45 kg (n=1,610),
Junior Midget - 10-13 years, 36 — 51.8 kg (n=1,489),
Midget - 11-14 years, 40.5 - 60.8 kg (n=1,160),

Junior Bantam - 12-15 years, 49.5 — 67.5 kg (n=177).
The coaches were the primary individuals responsible for reporting injuries, but
coaches and players were contacted at the end of the season to provide the
names of injured players. Only an injury that caused a player to miss at least
one week of participation was recorded and labeled as a significant injury.
Injuries were further categorized as moderate (8-21 days lost), major (>21 days
lost), and severe (permanent disability).

Overall, 257 injuries were reported (Goldberg et al., 1988). The oldest
and heaviest teams (Junior Bantam) had the highest injury prevalence (9.6%),
while the smallest and lightest teams (Junior Pee Wee) had the lowest
prevalence (1.9%). The authors examined several biological variables and their
association with an increased chance of injury. There was no significant
correlation between the age of players within a group and injury, but there was a
significant correlation between weight of players and injury only in the Midget

division.

30

The distributions of injuries in games and practices were 61.5% in games
and 30.7% in practice; the remaining injuries occurred during scrimmages
(7.8%). Scrimmages were not defined as inter-squad or intra-squad. Injuries
during scrimmages were considered separate from injuries occurring during
practice even though scrimmages usually take place during a practice session.
Players with the most playing time were injured more often; > players who
averaged playing at least 3 quarters incurred 90% of the injuries. Thus, as
exposure time increased, risk of injury increased. Even though exposure time
was estimated, exposure rates were not calculated.

Goldberg et al. (1988) only accounted for injuries that required cessation
of participation for at least 7 days. With this criterion for a reportable injury, many
injuries may go unnoticed, i.e., minor injuries requiring less than 7 days of
cessation would not be included. If a different definition of a reportable injury
were used, the injury rates in this study would probably be higher.

The method of injury reporting also has some limitations. Of the total
number of 257 reported injuries, 211 were reported by the coaches. The
coaches failed to report 46 injuries. Players reported 147 of the 257 total injuries
(Goldberg et al., 1988). This suggests different perceptions of injury between
players and coaches. For both coaches and players, end of the year reporting
may not be the most reliable method of data collection. Even though instructions
were given to the coaches on how to document injuries at the beginning of the

season, close to 20% of the total injuries went unreported by the coaches.

31

In an earlier study by Goldberg and colleagues (1984), 436 athletes were
studied in a prospective assessment of youth football injuries during the 1981
season in a Pop Warner youth football league in six New England towns. The
league was divided into three divisions using weight and age:

Pee Wee: 9-12 year olds, 65-100 lbs (29.5 - 45.5 kg),

Junior Midget: 10-13 year olds, 80-115 lbs (36.4-52.3 kg),

Midget: 11-14 year olds, 100-130 lbs (45.5-59.1 kg).

Questionnaires were completed by league personnel, and included information
on the diagnosis and site of injury, duration of disability, and age, weight and
position of the injured athlete. If a physician diagnosis was not available, the
researchers made the diagnosis based on a description of the injury. The
definition of injury depended on the length of time the athlete was unable to
participate. An injury that prevented participation from 1-7 days was categorized
as minor, and one that prevented participation for more than 7 days was
considered significant. There were three subcategories of significant injuries:
moderate (restricting an athlete from 8-21 days), major (restriction more than 21
days), and severe (permanent disability). The injury criteria used in this study
were established by NAIRS (National Athletic Injury/Illness Reporting System).

A total of 67 injuries were reported. The questionnaire completed by
league personnel accounted for only 30 of the injuries, while telephone interviews
with the families of 401 players reported 67 injuries. There was a significant
difference in total injuries and significant injuries among age/weight divisions.

The Peewee division had the fewest and the Midget division had the most

32

injuries. There was also a significant difference within divisions between heavy
and light athletes; heavier athletes showed an increased incidence of injury
(Goldberg et al., 1984).

There are several limitations with this study. Similar to Goldberg et al.
(1988), league officials failed to report all of the injuries. League officials
reported less than one-half of the injuries which were reported by the families of
the young athletes. Although 92% of the families were contacted, some
information could be lacking. The authors also made injury diagnoses based on
descriptions in telephone interviews. Sound practice suggests that a diagnosis
should only be made during a clinical examination where the physician or
certified athletic trainer has visual and palpatory contact with the athlete. Verbal
descriptions by a child, adolescent and/or parent may not be sufficient for an
accurate diagnosis.

Roser and Clawson (1970) conducted a prospective study of 2,079 boys
participating in the Seattle Junior Football Program during the 1968 season. The
program was organized onto five leagues (competitive divisions) with 70 teams
participating. The leagues/divisions were divided according to the player weight
and age:

Bee: 9-11 years old, up to 85 lbs (38.6 kg);

Midget: 9-11 years old, 86-98 lbs (39.1-44.5 kg), but teams could have five

12 year olds less than 75 lbs (34.1 kg);

33

Peewee: 10-12 year olds, 99-105 lbs (4547] kg), but teams could have

five 13 year olds less than 92 lbs (41.8 kg);

Gil Dobie: 13-14 year olds, 106-124 lbs (48.1-56.4 kg);

Bantam: up to 15 years old, 125-160 lbs (56.8-72.7 kg).

The season lasted 12 weeks. The first four weeks involved preseason practice,
2 hours per day, 6 days per week. After the first game, practices were reduced
to three, 2-hour practices per week. Each team played 7-8 games throughout
the season.

The data collection procedures involved a questionnaire which was
completed by the coaches every time an athlete was injured severely enough to
require missing a practice session or a game. The coaches were contacted
several times throughout the season to ensure that the data were being
collected. If an athlete was injured, the parents of that athlete as well as his
physician were contacted to gain additional information and a proper diagnosis.
At the end of the season, insurance companies were contacted to cross check all
medical claims (Roser and Clawson, 1970).

The results indicated that only 2.3% of the athletes sustained a football
injury severe enough to keep them out of at least one game or practice session.
Each player spent approximately 96 hours practicing and 12 hours in games; yet
the chance of injury in a game was 11 times higher than in practice. Lack of
heavy contact in practices and increased aggressiveness during games probably

contributed to this difference. Prevalence of injury varied by status: first-string

34

players, 81%; second-string players, 17%; and third string players, 2% (Roser
and Clawson, 1970). This finding probably reflected exposure time.

There was no association between age and the likelihood of injury in this
study. The Peewee division sustained the highest percentage of injuries (4.3%),
while the Gil Dobie division accounted for the lowest percentage (1.4%). The
Bee, Midget, and Bantam divisions sustained, respectively, 0.2%, 2.3%, and
3.3% of the injuries (Roser and Clawson, 1970).

Silverstein (1979) suggested that injury rates in youth football are higher
than those previously reported by Roser and Clawson (1970). Silverstein (1979)
examined youth football injuries for three teams over two seasons (1977 and
1978). The three teams represented three different levels of competition,

Peewee, Junior Varsity (JV), and Varsity (V):

Peewee: age 8-11 years, up to 90 lbs (41kg),
Junior Varsity: age 12-13 years, up to 110 lbs (50kg),
Varsity: age 14 years, up to 130 lbs (59kg).

The overall injury prevalence was 18.1% during the 1977 season and 12.9%
during the 1978 season. The Peewee team experienced an injury prevalence of
10.0% during the 1977 season and 14.4% during the 1978 season. Junior
varsity players had an injury prevalence of 28.3% during the 1977 season and
13.0% during the 1978 season. The Varsity team sustained an injury prevalence
of 20.6% in 1977 and 11.0% in 1978. The disparity in prevalence’s between
Junior Varsity and Varsity teams for the 1977 and 1978 seasons was attributed

to a low number of participants during the 1977 season.

35

There are very few studies which deal with body size or biological
maturity, and their possible relationships with injury in youth sport. It has been
suggested that body fatness may increase the risk of musculoskelotal injuries,
either directly or indirectly. Gomez et al. (1998) examined the relationship of
fatness and injury rates in high school football linemen. The sample included
215 junior varsity and varsity linemen who participated in ten high schools in San
Antonio, Texas. Athletic trainers employed at each high school collected the
data. Preseason evaluations of each lineman at the participating schools
included skinfold thickness at three sites (chest, abdomen, and front of the thigh),
and height and weight. All skinfold measurements were taken by an individual
experienced in the techniques. Body density was estimated using the equation
of (Jackson and Pollock, 1985):

Body density = 1.10938 — 0.0008627 (sum of 3 skinfolds) + 0.0000016 (sum of 3
skinfolds)2 — 0.0002574 (age).
Density was subsequently converted to percentage body fat using the equation
by Brozek et al (1963):
% body fat = {(4.57/body density} -4.412) x 100

A total of 67 linemen suffered 86 injuries over the course of the season, an
estimated injury rate of 5.66/1000 hours. Each school’s athletic trainer, who
recorded the data, estimated total exposure time per athlete. Total exposure
time was equated with total playing time and was calculated by estimating the
time each player spent physically participating in practices and games. The

schools trainers estimated that players spent 60% (1.2 hours) of each practice

36

actually engaged in physical activity. Players were credited with 6 minutes of
playing time per quarter if an athlete played one-way (offense or defense) and 12
minutes if they played both. Practice and game time were added together to
determine total exposure time for each athlete. There were no significant
differences between groups above, and at or below different levels of body fat
percentage. However, there was a significant difference in injuries to the lower
extremity by level of fatness. Players with an estimated fatness > 20% had a
significantly higher incidence of lower extremity injuries than those with body fat <
20%. However, a level of body fat (mean or median split, etc.) could not be
determined where there was a consistently significant difference in estimated
rates of injury between higher and lower fat groups (Gomez et al., 1998).
Limitations of this study are several. Error associated with the prediction
of body fat from skinfolds was not considered. Moreover, measurement error
tends to be greater in the case of larger skinfolds. Individual differences in body
fat, weight and the BMI were not considered in the analysis. Instead, injury rates
for specific BMI’s, weight, and body fatness were calculated. The study sample
was limited to high school football linemen, usually the largest athletes in the
sport. It was hypothesized by the authors that increased body fat was a risk
factor for injury in sports. As such, only the heaviest, and most likely fattest,
athletes were included in the study. Athletes playing other positions besides
offensive and defensive line were not included. Traditionally, skill positions
players and linebackers are leaner than linemen. It would have been helpful to

examine the weight, BMI, and percentage fat of all athletes.

37

Backous et al. (1988) estimated maturity status among male soccer
players and examined its relationship to injuries sustained during a one-week
soccer camp. Maturity status was estimated from grip strength and height.
Three maturity levels were established: immature, mature, and mature but weak.
Mature boys were > 165 cm in height and had a grip strength > 25 kg. Immature
boys were < 165 cm in height and had a grip strength < 25 kg. Some boys were
tall (above 165 cm) but did not demonstrate grip strength of 2 25 kg; these boys
were labeled mature but weak. Tall but weak boys had a higher incidence of
injuries than immature and mature boys. The authors assumed that tall, weak
boys may have greater sexual maturity than muscular development, but sexual
maturity status was not assessed. When comparing boys just on grip strength,
there was no difference in injury rates among strong and weak boys (Backous et
aL,1988)

Equating height and grip strength with pubertal maturity is a major concern
in this study. The authors stated that boys less than 165 cm with a grip strength
< 25kg would most likely be pubertally immature (Tanner stages 1-3). Specific
stages of pubertal development were not indicated, i.e. pubic hair or genital
development. Stages of sexual maturity described by Tanner (1962) are
characteristic specific and should not to be averaged or combined. Further, boys
in genital or pubic hair stage one are pre-pubertal, while boys in stages 2 or 3 are

already pubertal (Malina and Bouchard, 1991).

38

Age Grouping in Youth Sport

Participants in youth sports are grouped in several ways. The most
common method is by chronological age. Other methods are by grade in school,
ability, and physical maturity. Sometimes a combination of methods is used, for
example, if an athlete is in the appropriate by age, but exceeds the weight limit
for the age group, he could be moved to the next higher level. Pop Warner
Youth Football classifies children of similar age and size (Dick Butkus Football
Network, 2001). They are the only league that uses an age/weight matrix, which
is believed to reduce the number of injuries. Players who exceed a certain
weight limit must participate at the next level of competition.

The Mid-Michigan Pony Football League (MMPFL) uses a different
grouping method. Teams in the MMPFL are grouped primarily by grade in
school, but there are several age restrictions secondary to grade. The primary
safety rule used by the MMPFL is a weight restriction on primary ball carriers
(quarterbacks, running backs, receivers). Fourth and fifth grade teams cannot
use a player as a primary ball carrier who weighs more than 120 lbs (54.5kg) in
full football attire. Players on the sixth grade teams cannot use primary ball
carriers above 130 lbs (59 kg), and seventh grade teams above 140 lbs (63.5
kg). There is no weight restriction with the 8th grade team regarding primary ball
carriers. Players exceeding the weight requirement must play from tackle to
tackle (MMPFL, 2000). The rational for limiting the weight for ball carriers is that
a much heavier athlete will not be able to run into lighter athletes who are playing

defense. The rule that limits the weight on ball carriers does not apply for any

39

defensive position. Therefore, the heavy athlete can play defense and
presumably run full steam into a lighter ball carrier. Note, however, that boys 9-
14 years of age differ considerably in size and maturity (Malina and Beunen,
1996).

Assessing Maturity Status

Size differences among children and adolescents reflect, to a large extent,
individual differences in tempo of biological maturation, which are especially
pronounced during the transition into puberty. The biological status of an
individual can be assessed in several ways. These assessments include, but are
not limited to, skeletal maturity using hand/wrist radiographs, sexual maturity
using stages of secondary sex characteristics, dental maturity using dental X-
rays, somatic maturity using age at peak height velocity (PHV), percentage of
adult height, and perhaps muscular strength. Each method of maturity
assessment has limitations with regard to accuracy, practicality and cost.

The development of the skeleton spans the entire period of growth and is
perhaps the best bodily system for assessing maturity status (Malina and
Bouchard, 1991 ). There are three methods of assessment, which are commonly
used to estimate skeletal maturity: the Greulich-Pyle, Tanner-Whitehouse, and
the Fels methods.

The Greulich-Pyle method involves matching, as closely as possible, the
hand/wrist X-Ray of a child/adolescent with a series of standard hand/wrist X-
rays plates in an atlas. The method should be applied by assessing the maturity

status of each individual bone, but many studies simply compare the film of the

40

child/adolescent as a whole to the plates in the atlas (Malina and Bouchard,
1991)

The Tanner-Whitehouse method is sometimes called the bone specific
method. It entails matching the features of 20 individual bones of a hand/wrist X-
ray to written criteria for each bone. Each stage is given a specific score and the
scores for all bones are summed to give a maturity score. A skeletal age is then
derived from the summed score for 20 bones. Separate skeletal ages can be
assigned radius, ulna and short bones and to the carpals (Malina and Bouchard,
1991)

The Fels method also evaluates maturity indicators based on shape
changes and ratios between linear measurements. Grading is done by matching
the hand/wrist radiograph with the described criteria, and measurements of
epiphyseal and metaphyseal widths. The assigned grades are entered into a
personal computer, which calculates a skeletal age and the associated standard
error (Malina and Bouchard, 1991).

Even though skeletal age may be the best method of assessing maturity
during childhood and adolescence, there is little practicality for its use in studies
of large samples. The cost of the X-rays and the time it would take to assess the
films may be prohibitive. In addition, x-rays involve a low level of radiation
exposure.

Sexual maturity is based on the assessment of secondary sex
characteristics: breast and pubic hair development in girls, and genital and pubic

hair development in boys. Five stages of development are described for each

41

characteristic by Tanner (1962). Stage 1 is prepubertal; Stage 2 indicates the
onset or initial development of each characteristic; Stages 3-4 represent further
development; and Stage 5 indicates the adult or mature state. If a boy is
assessed as having stage 2 genital (G) and stage 4 pubic hair (PH), it is
expressed as G2 and PH4. Each characteristic should be treated separately,
i.e., not combined into a single score. Also, it cannot be said that a child in a
particular stage of development in one characteristic will automatically be in a
certain stage of development in another characteristic, i.e., PH 3 does not equal
G 3 (Malina and Bouchard, 1991).

Assessment of sexual maturity requires invasion of privacy. It should be
done in a clinical setting by direct visual observation. This poses a problem
because this type of assessment is invasive. It may be difficult to receive
consent from parents and children to perform this type of assessment. Another
limitation is that sexual maturity assessment is only of value only during the years
of puberty. Some boys in the age range 9-14 years will be still prepubertal.

Dental maturity assessment requires the use of dental X-rays to evaluate
the calcification of permanent teeth, but dental maturity proceeds independently
of skeletal, sexual, and somatic maturity (Malina and Bouchard, 1991). Dental X-
rays taken during a routine dental visit can be used to assess dental maturity.

Grip strength has been suggested as an indicator of maturity status and
as a pre-participation screening criterion in some sports. Backous et al. (1990)
suggested that grip strength combined with height is a good predictor of pubertal

maturity. The criteria used by Backous et al. (1990) to classify a boy as

42

“average” in maturity status were 25 kg of force using a hand-grip dynamometer
(Jamar, Asimow Engineering 00., Los Angeles) and a height of 65 inches (165
cm). A boy was considered immature if grip strength and height fell below 25 kg
and 65 inches (165 cm), and mature if they were above these levels. These
criteria were compared to stages of pubic hair: stages PH 1 through PH 3 were
considered immature, and stages PH 4 and PH 5 were considered mature.
Using height and grip strength, 70% of the boys were correctly classified as
mature or immature. Using grip strength combined with height might be an easy
method of assessing maturity, but a significant number of boys were
misclassified (30%). Further, defining PH 1 through PH 3 as immature is
somewhat naive, as is classifying PH 4 as mature. By definition, only Ph 5 is
mature (Tanner, 1962). Grip strength is also influenced by body size
independent of maturity status as well as the type of dynamometer used.

Assessing physical maturity in the preparticipation health evaluation may
be important for two reasons. It may help to reduce serious injuries to immature
athletes by limiting their participation in collision sports, and it may help to match
participants with respect to their abilities (Kreipe and Gewanter, 1985).
Nevertheless, data are lacking to suggest that the biologically immature athlete
sustains serious injury with greater frequency than the biologically mature athlete
in collision and other sports.

Somatic maturity is most often estimated as age at PHV, the maximum
rate of growth during the adolescent spurt. Estimation of age at PHV requires

longitudinal data, which limits its utility.

43

Expressing current height as a percentage of predicted adult height may
provide an estimate of somatic maturity. If the subject’s adult height can be
predicted, then current height can be expressed as a percentage of predicted
adult height and provide an indication of maturity status (Malina and Bouchard,
1991; Malina and Beunen, 1996). Data from longitudinal studies indicate that
within an age group (e.g., 10 years), children advanced in maturity attain a
greater percentage of adult height compared to average and late maturing
children (Bayer and Bailey, 1959).

Prediction of adult stature usually requires a skeletal age assessment, in
addition to age, weight, height and midparent height. Hand/wrist radiographs are
not readily available in surveys and when they are available, relatively few
individuals have the training to accurately assess them. A height prediction
method has been developed that can be used in the absence of skeletal age to
facilitate prediction of adult stature (Khamis and Roche, 1994). There are four
predictor variables needed with the Khamis-Roche prediction method: current
age, stature and weight of the child, and midparent stature. The regression
equation has the following form within an age and sex group:

predicted adult height = 60 + 81 stature + 62 weight + [33 midparent

stature,
where [31, [32, 83 are the coefficients by which stature, weight and midparent
stature should be multiplied. Coefficients change with each half-year of

chronological age. The method decreases in accuracy somewhat when applied

44

to boys around 14 years of age due to the omission of skeletal age (Khamis and
Roche,1994)
SUMMARY

Studies of sports injuries use different methods of data collection and
definitions of injury. Both undoubtedly influence estimated injury rates. Tables 2
and 3 summarize the trends in the available studies of football. It is clear that
some definitions of a reportable injury are more ambiguous or less strict than
others, resulting in estimated injury rates ranging from 10.5% to 81.1% in high
school football.

Data collection procedures can also have an impact on injury rates.
Retrospective surveys and data obtained from insurance forms and/or hospital
records show lower injury rates than direct, on-site interviews, or daily reports.
According to Thompson et al. (1987), direct interview techniques offer the best
source of incidence data, regardless of the scope of the study or definitions used
for injury. Only two of the studies cited (Powell and Barber-Foss, 1999; DeLee
and Farney, 1992) used certified athletic trainers to collect data on high school
football injuries.

Much of the literature on youth sports includes data for an assortment of
sports and activities, which makes it difficult to establish injury rates for specific
sports. Large-scale surveys of sports injuries to children and adolescents
adequately show how many people are injured, but risk factors are not taken into

account

45

Quality data on football injuries in community level youth programs are
lacking. The accuracy of reported injuries has limitations. Retrospective
telephone interviews require the parent or child to recall an injury event that may
have occurred months ago. Other methods require a coach, league official, or
team representative to collect and report injury data. These individuals may be
lacking in medical training and also are busy with other responsibilities during
games or practices. Regardless of instructions from researchers, uniform
educational background on sports medicine injuries would be a benefit. The
present study used a strict operational definition of a reportable injury and
certified athletic trainers to collect injury data on a daily basis. Combining both of
these methodological techniques should provide more accurate estimates of
injuries in youth football.

There are several methods of estimating biological maturity status:
skeletal age, secondary sex characteristics, dental maturity, somatic maturity
using age at peak height velocity (PHV), and percentage of adult height. Skeletal
age and sexual maturity have been used widely in growth and performance
studies of young athletes (Malina, 1994, 1998), but have not been used on a
regular basis in the study of injuries (Malina, 2001). Percentage of predicted
adult height has apparently not been used to estimate maturity status of youth
football players and has not been evaluated as a possible risk factor for injury in
any youth sport. This method of estimating maturity and examining its possible
relationship with injury is youth football is the most practical for this type of study.

Estimating predicted adult height as a method of assessing maturity is non-

46

invasive and requires only basic information from the subjects (height, weight,

age, heights of biological parents).

47

CHAPTER 3
METHODS

The purpose of this study was to examine the prevalence and incidence of
injury and the risk of injury in youth football participants. lntemal variables, or
player related risk factors, were considered, and their relationships with risk of
injury in youth football were examined. The data for this study were from the first
year of a two-year project funded by the National Athletic Trainers Association
Foundation with Robert M. Malina as the primary investigator. The study was
approved by University Committee on Research Involving Human Subjects
(UCRIHS) at Michigan State University (Appendix I).
Population

The subjects were 355 youth football players 9-14 years of age in the 4th
to 8th grades from two communities in south central Michigan (Holt [n = 210] and
St. John’s [n = 145]). The criterion for inclusion was registration with the youth
football program in the community and membership on a team. In addition to
permission from league officials, informed consent from both parents and child
were obtained from 296 participants. The informed consent (Appendix II)
allowed use of the biological parents’ heights in order to predict the adult height
of the athlete. There were 58 participants whose parents did not give consent for
the study and, therefore, heights of the biological parents were not available.
Although permission was not granted, these 58 boys were included in the
exposure and injury statistics as per agreement with the league. Their heights

and weights were also measured as per league requirement and request.

48

These boys, however, did not complete the questionnaire to assess perception of
risk of injury.
Project Design

This is a prospective cohort study. Data were recorded as events
unfolding during the season with few exceptions, such as injuries occurring at an
away game (each team played four away games). The prospective nature of the
study was not compromised even though each team played four away games
that were not covered by an athletic trainer. Any injury sustained during an away
game was investigated within a day after the game. Even if an injury surveillance
study adheres to the strictest definition of being prospective, there are always
some injuries that are not brought to the attention of the investigator immediately
after they occur. Some athletes may sustain an injury and may not reveal the
injury for days or weeks after the initial incident.

Cohort models are better suited for sport level analysis and for estimating
injury rates. The case-control method was also used. Case-control models are
better suited for player level analysis, deriving odds ratios (OR) and injury
prediction.

The most important aspect of any study of injury in sport is the operational
definition of an injury. Several definitions have been adopted during the history
of injury surveillance. It is difficult, if not impossible, to compare injury studies
when the definition of a reportable injury differs from study to study. The

following definition of a reportable injury (Powell and Barber-Foss, 1999, pp. 278)

49

has been used in high school injury surveillance studies and was used in the
present study:
0 “Any injury that causes cessation of participation in the current game or
practice and prevents the player’s return to that session
0 “Any injury that causes cessation of a player’s customary participation on
the day following the day of onset.
c “Any fracture that occurs, even though the athlete does not miss any
regularly scheduled session
0 “Any dental injury, including fillings, luxations, and fractures.
c “Any mild brain injury that requires cessation of a player’s participation for
observation before returning, whether in the current session or the next
session.”
Since there were certified athletic trainers on site at each practice and home
game in the present study, an operational definition of a reportable injury with this
detail was applied. With retrospective data or data collected without on-site
personnel, this definition of a reportable injury would not be appropriate.

Severity was also based on the model of Powell and Barber-Foss (1999).
Injury severity categories were labeled as minor (1-7 days lost), moderate (8-21
days lost), and major (>21 days lost). If an athlete sat out one day for an injury,
returned to competition/practice, then sat out another day for the same injury, he
was counted as missing two days. If an athlete missed one day for an injury,
returned to competition/practice, then missed one day for a new injury, the

athlete was counted as missing one day for each injury.

50

Data Collection

The heights and weights of the players were measured at the beginning of
the season (August). Boys wore shorts and a t-shirt, and shoes were removed.
Height was taken with a field anthropometer (GPM, Martin type, Pfister Import-
Export, Inc.) to the nearest millimeter, and weight was taken with a digital scale
to the nearest 0.2 kg. The digital scale was checked at the beginning of every
weighing session to make sure it started at zero. For height, the boy was
standing erect without shoes and the field anthropometer was placed behind him
and aligned with the spine. Replicate measurements were taken on 29 boys.
The technical error of measurement (oe) was calculated as the square root of the
sum of differences between replicate measurements squared divided by twice
the number of replicates (Malina, 1995):

oe = ‘1 Z d2/2N.
The intraobserver technical error for height was 0.22 cm.

The body mass index (BMI) was calculated as weight (kg) divided by
height (meters) squared (kg/m2). The BMI is an indicator of heaviness, not
necessarily fatness in children and adolescents (Malina and Katzmarzyk, 1999).

The leagues established the grouping of participants. League enrollment
was open to any child/adolescent in the two communities who was in the 4th
through 8th grades. There were two methods of grouping, which were
determined by the league (MMPFL, 2000). The teams were first grouped by
grade. For example, anyone enrolled to participate in the football league and

was in the 6th grade, played on the 6th grade team regardless of age. If a

51

student was held back a year, or advanced a year in school, he participated with
the grade in which he attended in school. The second method of grouping only
applied if there were enough participants in a particular grade to form several
teams. At the very beginning of the season all participants who were in the same
grade practiced together. The coaches then had a “draft” and chose players for
their respective teams.

The football season lasted from mid-August until the end of October. The
first four weeks consisted of practice four times per week. It was mandatory for
each player to undergo 8 hours (4 practice sessions) of conditioning before
contact drills could be performed. If a player registered late, he still had to
undergo the 8 hours of conditioning even if the rest of the team was practicing in
full pads. Once games began, practice was reduced to three times per week,
except for the eighth grade team, which was allowed to continue practicing four
times a week. Each practice session lasted 2 hours. During this time, the
athletes did a warm up consisting of a light jog, followed by stretching. After the
warm-up, coaches had the players run through drills such as tackling, blocking,
and agility. Each team had enough players to field a full offense and defense.
The practices usually ended with the teams running plays.

Each team played six games over the seven-week season. A NATA
certified athletic trainer (ATC) was on site to record the number of participants at
all practices and games, i.e., coach-directed sessions which were opportunities
for injury (exposures) and injuries as they occurred. The ATC was the primary

on-site caregiver to injured athletes (as agreed by the league organizers) and the

52

primary recorder of injury data. If an athlete was injured during a practice
session or a game, the ATC recorded the athlete’s grade, position or activity at
the time of injury, the assessment of the injury (body part, type of injury),
perceived severity, weather and field condition, and the action taken - return to
participation, removal from participation, taken to hospital (Appendix III). With
regard to away games, the ATC consulted with the head coach at the next
practice session to inquire about injuries that may have occurred during the away
game. If the coach indicated that an athlete did sustain an injury at an away
game, the athlete was then interviewed by the ATC. The certified athletic trainer
talked to the parents if the injury was perceived to be serious enough to warrant
being seen by a family or emergency room physician. If a child did seek the
attention of a physician, further contact was made with the parents of the injured
child to attain a more detailed description of the diagnosis. The ATC inquired
about the context of the injury, the body part involved, and any time loss. The
ATC performed an evaluation of the injury and contacted the parents to discuss
possible treatment options. Along with recording injury data, daily logs of
attendance, type of session, activities carried out during the session, and
weather conditions were recorded (Appendix IV). This information was used to
estimate exposure and related conditions.

Participants were also asked to complete a 24-item questionnaire
(Appendix V) dealing with the perception of risk of injury in sports (Kontos et al.,
no date). The questionnaire was administered to the athletes during a practice

session. Athletes, who were absent from practice when the questionnaire was

53

issued to their team, missed the opportunity to complete it. Efforts were made to
track down athletes who were absent, but compliance was not high. Of the total
sample, 240 (72%) boys returned completed questionnaires.

The Risk of Injury Sports Scale (RISSc) requires the participants to
answer a series of questions on a six-point scale, 1=very unlikely, 2=unlikely,
3=somewhat unlikely, 4=somewhat likely, 5= likely, 6=very likely. All questions
began with the following: What do you think are the chances that you will...? An
example of 1 of the 24 possible completions to the beginning of the question is:
injure yourself in a collision with an opponent?

Predicting Adult Height

Percentage of predicted adult height was used to estimate biological
maturity status. The procedure is based on the assumption that the closer an
individual is to his adult height, the greater his level of maturity (Malina and
Bouchard, 1991; Malina and Beunen, 1996). The prediction equation used in the
study required the current age, height, and weight of the boy, and midparent
height of his biological parents.

Reported heights of both biological parents were obtained for 296
athletes. Since individuals ordinarily over estimate self-reported heights, a
correction for reported heights was used (Epstein et al., 1995). The correction
formula for males was:

2.316 + (0.955 * reported height in inches),
and for females was:

2.803 + (0.953 * reported height in inches)

54

Midparent height, the average of parental heights, was calculated as follows:

corrected height, father + corrected height, mother
2

The chronological age of each athlete was converted to a decimal age at
the time of examination. This is done by subtracting date of birth from date of
examination.

The Khamis-Roche Method for predicting adult height was used. The
method requires the child’s, current age, height and weight, and midparent
height. The formula used for predicting adult heights is as follows:

(30 + (stature * stature constant) + (weight * weight constant) +
(midparent height * MPH constant),
where height is in centimeters, weights in kilograms, and MPH in centimeters.
The intercept and the constants change for each half-year of decimal age
(Khamis and Roche, 1994).

A median absolute deviation (MAD) was used by Khamis and Roche
(1994) at each chronological age to measure the accuracy of the prediction
method. The MAD is the "...median of the absolute values of the differences
between the actual 18 year statures and the predicted 18 year statures” (p. 505).
The MAD for the Khamis-Roche (KR) method was compared to the modified
RWT (Roche, Wainer and Thissen) method of height prediction, which
incorporated skeletal age in the formula. The 90% error bounds for the KR
method slightly exceeded those for the modified KWT method by about 0.3

inches or 0.76 cm in males (Khamis and Roche, 1994).

55

Predicted adult heights were estimated for 36 of the 58 subjects for whom
parental data were not available. Mean midparent height of the population was
used in the equation. Adult height was not predicted for the remaining 22 boys
because one or more of the necessary variables for the calculation were not
recorded (age, height or weight).

Predicted adult height and percentage of predicted adult height was
estimated for 36 participants for whom parental heights were not available. The
mean midparent height of the population (Table 4) was used in the prediction
equation. The mean percentage of predicted adult height was 84.2% in boys for
whom parental heights were reported with a range of 74.2% to 95.9%. In the
sample for which mean midparent height was substituted, the mean percentage
of predicted adult height was 83.6% with a range of 74.0% to 94.8%. There does
not appear to be a difference between the samples.

After the predicted adult stature was derived for each athlete, maturity
status was estimated by expressing current height as a percentage of the
predicted adult height. This was done by dividing current height by predicted
adult height and expressing it as a percentage. The derived variable is
confinuous.

The boys were also divided into maturity categories. An athlete whose
percentage of predicted adult height was > 1 SD above the mean-for-age
reported by Roche et al. (1983) for the Fels longitudinal sample was considered
advanced (early) in maturity, and an athlete whose percentage of predicted adult

height was >1 SD below the mean-for-age was considered delayed (late) in

56

maturity status. An athlete who fell within +/- 1 SD of the mean for age were
considered average or “on time” in maturity.

Roche et al. (1983) report means and standard deviations for single age
groups from 5 to 15 years; half-year estimates were not reported. Children in the
Fels study were seen within one month of their birthdays. For classifying youth
football players in this study as early, average or late in maturity, the Fels means
and standard deviations closest to the age of each football player were used.
For example, boys between 10.50 and 11.49 years were considered 11 years
old, and the mean and standard deviations for 11-year-old boys in the Fels
sample was used.

There are very few published articles examining any type of biological
maturity indicator and the risk of injury in youth sports. The method of predicting
adult stature to assess maturity status has apparently not been used in a sports
injury study. Because the Khamis-Roche method is non-invasive, it may be a
practical method of estimating the maturity status of children and adolescents.
Injury Rate Analysis

Along with injury information, the ATC recorded exposure data for each
grade. For the purposes of the study, an exposure occurred any time an athlete
participated in a game or practice session and had a chance of injury. League
rules required that a player participating on a 4th-7th-grade team play at least 6
plays in each half of a game. There were no set minimum of plays for 8th grade
players. Exposure data were recorded for each practice and each game (home

and away). Exposure data allowed for the calculation of injury rates per Athlete

57

Exposures (AE). Exposure rates were calculated for the entire population, each
community, and each grade, and were also differentiated between practice and
game rates. The formula for calculating exposure rates was as follow:

(total injuries / athlete exposures) x 1000,
where exposure rate is expressed as injuries per 1000 AE (Powell and Barber-
Foss,1999)

Player rates and case rates were calculated for the population as a whole,
each community, and for each grade. These were expressed as player rate/100
players and case rate/100 players. These rates are different from each other
because case rates include multiple injuries (Powell and Barber-Foss, 1999).
Case rates will always be equal to or higher than player rates. The formulae for
calculating player rates and case rates were as follows:

Player rates = # of athletes injured at least once / total # of players

Case Rates = # of total injuries / total # of players
Statistical Analysis

Descriptive statistics were calculated for age, height, weight, the BMI, and
percentage of predicted adult height by whole year age group, i.e., 10.0 to 10.99,
11.0-11.99, etc., and by grade. Descriptive statistics were established for each
community, each grade/division within each community and each grade/division
combined. Age-specific means and standard deviations for height, weight, and
the BMI were compared to reference values for American boys (Centers for
Disease Control and Prevention, 2000). Age-specific means and standard

deviations for the percentage of predicted adult height were compared to the

58

corresponding means and standard deviations for boys in the Fels sample
(Roche et al., 1983). After describing the characteristics of the sample,
descriptive statistics on injury type, location, position, and rates were calculated.

Chi square (x2) analysis was used to compare actual and expected injury
distributions. Height, weight, BMI, and maturity status were divided into tertiles
as well as a mean split within grade for these analyses. These variables were
also cross-tabulated with injured and non-injured players. Tertile categories were
established based on the study population. Tertile categories based on the study
population rather than national reference data was more appropriate for 2
reasons. First, the study population is not-representative of the national
population. Participants in a football league are not a random cross-section of
the population. Secondly, there will be too few subjects representing the highs
and lows for each variable. For example, if the BMI were categorized into light,
average, and heavy according to national reference values, there would be very
few subjects in the light range, and an excess number of boys in the heavy
range. Grouping into tertiles according to the population allowed a more equal
distribution of subjects. Even though a subject might be considered light or short
by national reference values, the subject may fit into a different category within
the population of youth football players.

The case control method focused on an injured player (case) and three
randomly selected teammates (controls) who were participants in the
game/practice at which the injury occurred. The cases (injured players) and

controls (non-injured players) were compared. Body size and maturity status

59

were split into high and low groups using appropriate methods (i.e., median split,
mean split, quartile split) in order to calculate the odds ratio (OR). Odds ratios
were calculated for body size (height, weight, BMI) and maturity status. The 95%
confidence interval was estimated for each OR. Odds ratios were compared
using a Mantel-Haenszel x2 test to determine the association between each risk
factor and injury. This OR was calculated by dividing the sum, over all strata of
XeYo/N by the sum, over all strata of YeXo/N. The Mantel-Haenszel OR is
different than a normal OR in that it is a precision-weighted average of the
stratum-specific estimates of the uniform OR (MacMahon and Trichopoulos,
1996). The OR takes values between zero (’0’) and infinity. One ('1') is the
neutral value and means that there is no difference between the groups
compared; close to zero or infinity means a large difference. An OR larger than
one means that group one has a larger proportion than group two, if the opposite
is true the OR will be smaller than one. If you swap the two proportions, the OR
will take on its inverse (1/OR) (SISA, 2001).

Because the body size and maturity variables are continuous, logistic
regression was used to assess their relative and combined contribution to the
prediction of injury in youth football. Linear regression deals with finding a
function that relates a continuous outcome variable (dependent variable) to one
or more predictors. Logistic regression is a variation of ordinary linear regression,
useful when the observed outcome is restricted to two values, which usually
represent the occurrence or non-occurrence of some outcome event, in this case

injured or non-injured. It was hypothesized that the predictor variables will

60

interact with each other to influence injury. Therefore, within the logistic
regression model, two-way and higher order interactions between the predictor
variables and injury were evaluated.

Answers from the RISSc questionnaire were grouped into 6 factors:
uncontrollable, controllable, overuse, upper-body, surface related, and re-injury
(Kontos et al., 2000). A mean score for each factor was calculated. After mean
scores for each factor were calculated, logistic regression was used to predict
injury, since injury is a binary outcome variable. Descriptive analysis and logistic
regression results were compared to results from Kontos (2000) to examine
similarities and/or differences in RISSc data among youth football players and
youth soccer players. Table 17 shows the reliability analysis of the RISSc

questionnaire.

61

CHAPTER 4
RESULTS

The present study is set in the context of four questions related to injury in
youth football:

1. What are the prevalence and incidence of injury in youth football?

2. Do injury rates differ by grade?

3. Are there biological variables (weight, height, biological maturity) that
are associated with the risk of being injured, and are there interactions among
these variables?

4. Is the perception of risk of injury in youth football related to the
likelihood of injury?

Each of the questions is subsequently addressed after the descriptive
characteristics of the sample are considered.
Descriptive Statistics

Descriptive statistics for the participants from the two communities and
total population are given in Table 5. Data are separated by grade because
teams in the Mid-Michigan Pony Football League (2000) were formed according
to current grade in school. There are, on average, no significant differences in
age, height, weight, the BMI, and the percentage of predicted adult height
between players in the two communities. Because participants were grouped by
grade in school, there is about a 2-year difference between the youngest and

oldest player on each team (Table 4).

62

Descriptive statistics for height, weight, and the BMI for the combined
sample of the two communities grouped by age are shown in Figures 1-3 relative
to the reference values for United States boys (Center for Disease Control and
Prevention, 2000). The youth football participants in the two mid-Michigan
communities are taller (at the 75th percentile) and heavier (above the 75th
percentile) than the general population of American boys of the same age. They
in turn have a higher BMI, which exceeds the 75th percentile of the reference
data

The mean percentage of predicted adult height of the football players is
compared to mean values for boys from the Fels Longitudinal Study in Figure 4.
On average, the percentage of predicted adult height in the youth football sample
attained at each age is consistent with the Fels sample. There is more variation
among the football players because they are grouped within whole year age
categories whereas the boys in the Fels sample were seen within one month of
their birthdays (Roche et al., 1983).

The athletic background of the sample is varied. The subjects had played
anywhere from none to 6 other sports besides youth football, with 3 being the
median number of sports (Table 6). The most frequent first sports played were
tee ball (35.5%) and then soccer (31.0%). Only 6.6% of the sample population
reported football as the first organized sport experience (could have been either
tackle or flag football).

Almost 30% of those responding to the of sport background questionnaire

reported sustaining a previous injury. The detail in the injury description varied

63

greatly among the reports as no operational definition of an injury was indicated
in the questionnaire.
Question 1: What are the prevalence and incidence of injury in youth football?

During the course of the season, 100 boys (28%) sustained a reportable
injury. Chi square analysis of injuries between the two communities showed no
significant difference (p = .254, x2 = 1.353) in the prevalence of injuries (30% and
26%). The prevalence of reportable injuries increased with grade level and to
some extent with age (Table 7). The older population (7th and 8th grade)
sustained significantly more injuries (p = .000, x2 = 21.962) than the younger
population (4th-6th grade) of youth football players.

Frequencies of injuries (including multiple injuries to an athlete) by type
are summarized in Table 8. Of the 100 boys who were injured, 31 boys
sustained a second injury and 6 boys sustained a third injury resulting in a total
number of reportable injuries to 137. The most common injury was general
trauma (41%), followed by strains (14%) and sprains (13%). There were 12
cases of neurotrauma (concussions, 9%). Over two-thirds (69%) of the injuries
were graded as minor (missing less than 7 days), 10% were graded as moderate
(missing 8-21 days), and 13% were graded as major (missing more than 21 days
and including 10 fractures regardless of time loss). Grading of the remaining 8%
of injuries was incomplete. In some cases the athlete never returned to play after
sustaining an injury and time loss was unable to be established, or athletes were
cleared to play, but did not by their own choice or their parents. Seventh grade

subjects sustained 4 of 9 fractures and 4 of 9 neurotraumas (concussions). Sixth

64

and eighth grade subjects sustained most of the general trauma injuries (11 and
13 of 42 injuries, respectively).

Reportable injuries by position or activity at the time of injury are
summarized in Table 9. Offensive linemen (tackle to tackle and tight end,
excluding split ends, flankers, wide receivers), defensive linemen (ends, tackles,
nose guards), and running backs accounted for more than one-half (53%) of the
reportable injuries.

Question 2. Do injury rates differ by age and grade?

Injury rates based on athlete exposures are summarized by community
and grade for practices, games, and both combined in Tables 10 -14. Exposures
were coach-directed sessions, i.e., practices and games (Table 10). Injuries
were more common in games (Table 12) than in practice (Table 11), although the
ratio of exposure was about 4 to 1 for practices versus games. The differences
in estimated injury rates for practices and games between communities across
grades were small, 12.1 and 9.7 per 1000 AE for communities A and B,
respectively. However, there was variation by grade between communities.
Sixth grade players from community A had higher estimated practice, game, and
overall rates than 6th grade players in community B, whereas 7th grade players
in community B had higher estimated practice, game and overall rates than 7th
grade players in community A. Estimated injury rates (per 1000 AE) for grades
and communities combined were 9.6 for practices and 18.0 for games (Table
14). The overall injury rate for practices and games together was 11.1 per 1000

AE.

65

Injury rates during practice sessions were two times higher among 7th and
8th grade participants compared to 4-6th grade participants. Injury rates during
games increased with grade level from 4/5th to 7th grades; the lower game rate
among 8th grade participants probably reflects sampling variation (only one of
the two communities had an 8th grade team). The incident density ratio (game
rate/practice rate) was highest for the 61h grade (2.11), identical for the 4/5th and
7th grades (1.65) and lowest for the 8th grade (1.28). The injury rate per 1000
AE was more than two times higher for games than practices among 6th grade
players.

For practice sessions and games combined, estimated injury rates differed
only slightly between 4/5th -6th grade participants and between 7th - 8th grade
participants. Overall injury rates for practices and games combined in 7th and
8th grade players were about twice as high as in 4/5th and 6th grade players.

Table 15 shows cases rates and player rates for each community and the
total sample. Table 16 shows player rates and case rates for each grade. Both
player and case rates increase with grade. As expected, case rates were higher
than player rates because case rates count the total number of injuries and
player rates only count the number of injured athletes. Players who sustained
more than one injury were counted only once in player rate analysis, yet each
injury was used to calculate case rates.

Question 3. Are there biological variables (weight, height, biological maturity)
that are associated with the risk of being injured, and are there interactions

among these variables?

66

In order to calculate the odds ratio (OR) to compare injuries at the player
level, height, weight and the BMI were split into two equal groups at the mean
(high and low). Each of the three variables was plotted on a 2 by 2 table and the
OR was calculated. Results are summarized in Table 17.

Odds ratios were calculated by using a mean split for height, weight and
BMI for each grade. Lower values (short, light in weight, and low BMI) had a
higher relative risk of injury or OR than the upper values in all cases and grades
except the 7th grade. Seventh graders, who were taller, had higher BMls and
higher weights were nearly 1.5 times more likely to sustain an injury than seventh
grade players who had lower values in height, weight and the BMI. The only
significantly high OR was between 4/5th grade boys with a low and high BMI
where lower BMI athletes were 3.18 times more likely to sustain an injury than
players with a higher BMI (p = .021, x2 = 5.340). The next group with the highest
relative risk was light 4/5th grade players. They are 2.5 times more likely to
sustain an injury than heavier 4/5th grade players.

In order to evaluate differences between physical characteristics and
injuries, participants within each grade were divided into tertiles for weight, height
and BMI. Percentage of predicted adult height was used to divide the sample
into three contrasting maturity groups - early, on time, late.

Chi-square analysis of weight, height, the BMI, and estimated maturity
status for injured versus non-injured players were run separately for each
variable. The chi-square for height, weight and BMI were split into tertiles based

on the study population, as well as mean splits within grade. Maturity was

67

divided into tertiles based on the Fels reference data. Results showed that there
were no statistically significant differences in injury distribution among players in
the tertiles when the population was split by to grade. However, when 4-6th
grade subjects were grouped together, subjects with low BMI sustained
significantly more injuries than subjects with medium and high BMls (p = .038,
x2 = 6.546) and less mature subjects sustained significantly more injuries than
on-time and early maturers (p = .009, x2 = 9.420). Older subjects (7-8th grade)
combined together did not show any significant differences in injury distribution
between low, medium and high categories of height, weight, BMI and estimated
maturity.

Chi-square analysis was run with the sample split into younger (4-6th
grade) and older (7-8th grade) grades with the biological variables (height,
weight, BMI) divided by a mean split into high and low values. A significant
difference in injury distribution between high and low values of BMI was seen in
the younger group. Lighter subjects were injured more than heavier subjects (p =
.015, x2 = 5.920).

When the population was split according to grade, and height, weight and
BMI split into high and low values, chi-square analysis showed that 4/5th graders
with a low BMI value sustained more injuries than those with higher BMl’s
(p = .021, x2 = 5.340). There were no other significant differences in injury
distribution with any other grade level and variable.

A backward stepwise logistic regression was run to analyze possible

predictors to injury. Backward stepwise logistic regression begins analysis with

68

all of the desired variables and then removes them one by one if they do not
contribute enough to the regression equation. For this analysis, the dependent
variable of injured versus not injured was run with height, BMI, and percentage of
predicted adult height as the independent biological variables along with the six
RISSc factors (uncontrollable, controllable, upper extremity, surface related,
reinjury and overuse) as independent psychological variables. Each independent
variable was arranged into tertiles. Because of grade related differences in injury
rates, separate analyses were done for subjects in the 4th-6th grades and 7th-
8th grades.

In the younger population (4th -6th grade), the BMI was the only
significant predictor of being injured. The final model showed that an increase in
the BMI was protective in nature (Table 19). The BMI was associated with an
OR [EXP (B)] of .426 (p = .044). In the older population (7th-8th grade), the final
model (Table 20) showed that higher concern for uncontrollable injuries was
protective in nature (p = .040), and higher concern for being reinjured increased
the risk of injury (p = .002). The OR [EXP (B)] for the RISSc factor labeled
uncontrollable was .438 and the odds ratio for the reinjury RISSc factor was 3.88.
Question 4. Is the perception of risk of injury in youth football related to the
likelihood of injury?

Partial correlations were run for each biological variable (height, weight,
BMI, maturity) with the six RISSc factors (uncontrollable, controllable, overuse,
upper extremity, surface related and reinjury). There were significant negative

correlations with each biological variable and the uncontrollable RISSc factor in

69

the older population (Table 21). Perception of risk of an uncontrollable injury

increased with an increase in height, weight and the BMI. There were no

significant correlations in the younger population.

The results of the study can be summarized in the context of the four

hypotheses generated:

1.

Injury rates increase with grade. This hypothesis was accepted. Injuries in
youth football increased with grade level of the competitors.

Weight, height, the BMI, and biological maturity by themselves do not
increase the likelihood of injury in youth football. This hypothesis was also
accepted. Weight, height, the BMI, and estimated maturity status were

not significant risk factors for injury in youth football.

. Player’s perception of risk influences the likelihood of injury. This

hypothesis was partially supported, but only in older players (7th and 8th
grades). Two expressed concerned related to perception of risk were
significant predictors of injury: uncontrollable factors and re-injury.

Internal variables (player-related factors) interact to influence the risk of
injury in youth football. This hypothesis was not supported in the analysis.

There were no significant interactions terms.

70

CHAPTER 5
DISCUSSION

Youth football is a popular sport. Because football is a contact sport, it
often results in injuries to participants. It is also a sport that is subject to scrutiny
regarding safety and the appropriate age, size and maturity of youth participants.
The present study is set in the context of four questions related to injury in youth
football:

1. What are the prevalence and incidence of injury in youth football?

2. Do injury rates differ by age and grade?

3. Are there biological variables (weight, height, biological maturity) that
are associated with the risk of being injured, and are there interactions among
these variables?

4. Is the perception of risk of injury in youth football related to the
likelihood of injury?

The results of this study have relevance to several areas. First, there is
limited data on injuries in youth football. This study provides in-depth analysis of
injury rates in present day youth football in two communities. It thus addresses
issues at the local level of youth sports, a level that has received much less
consideration than the interscholastic level. The most recent study on injuries in
youth football is that of Goldberg et al (1988). None of the previous studies on
injuries in youth football employed certified athletic trainers to record injuries on a
day-to-day basis. The prospective nature of this study is unique and important

especially with regard to accuracy in time loss from participation and the follow

71

up of injuries. The results of this study have the potential to build a base for
subsequent studies. The approach used in the present study provided sound
data. In addition, having a certified athletic trainer on hand for practices and
games permitted on field care of injuries as they occurred.

From a second perspective, player-related risk factors for injury in youth
football have not been systematically evaluated. The dissertation by Violette
(1976) was one of the only studies to evaluate biological maturity status in the
context of interscholastic football injuries. However, risk analysis using this
variable was not done. Violette assessed maturity with a composite score of
three secondary sex characteristics — testicular volume, pubic hair and axillary
hair. Estimating biological maturity in this manner is not practical for several
reasons. A trained individual is needed to assess secondary sex characteristics.
It would be best for one individual to assess the secondary sex characteristics,
but it would take too much time. Having several people perform the assessment
(such as physicians performing pre-participation physical examinations) would
decrease the reliability. Also, in today’s social climate, it would be very difficult to
gain parental consent to perform such an invasive assessment on a child or
adolescent. Therefore, a non-invasive method of estimating biological maturity
could be of value. The method of estimating biological maturity in this study has
apparently not been used previously in this context. If youth football leagues
were concerned about matching participants by maturity status, coaches or

league officials could easily utilize the non-invasive method adopted in this study.

72

The study design could be modified to examine body size and maturity as
risk factors for injury in other sports, as well as for the estimation of injury rates.
Height and weight are relatively easy measurements, and the BMI is a rather
simple calculation. In fact, youth football leagues usually require height and
weight, and coaches and league officials will most likely be happy to have a
researcher take the heights and weights.

A comparison of this study to previous research is difficult. There are no
previous studies that use similar study design, definition of a reportable injury, or
population in examining injuries in youth football. A three-year study conducted
by the National Athletic Trainers Association (Powell and Barber-Foss, 1999)
examined 10 high school varsity sports, including football. The exposure criteria
and definition of a reportable injury were similar to this study with the main
difference being the age of the players. For both studies, injury rates were higher
in game sessions than during practices. Game rates were four times higher than
practices in the NATA 3-year study and were about twice as high in this study. In
both studies, most injuries were classified as minor (loss of 1-7 days). High
school players in the NATA 3-year study suffered more sprains than any other
type of injury (31.7%), while youth football players in this study suffered more
general trauma injuries than any other (40%), with sprains ranking third with
13.1% of the injuries. Injury rate comparisons between the youth football players
in this study and the high school players in the NATA study revealed that high
school football players have higher case and player rates yet the youth football

players had a higher exposure rate.

73

There are several published studies on injuries in youth football and even
though injury rates in those studies do not resemble the injury rates of this study,
there are some similar trends. The study by Goldberg et al. (1988) examined six
New England Pop Warner Football Leagues and established that the older,
heavier players had the highest injury rates and the smaller, lighter teams had
the lowest injury rates. Similar results were found in an earlier study by Goldberg
(1984); however, a study by Roser and Clawson (1970) did not show this trend.

Another similarity between this study and the study by Goldberg et al.
(1988) is the risk of injury between practice and game sessions. Although
Goldberg and colleagues did not estimate exposure rates, they established that
twice as many injuries occurred during games than in practices, much like this
study.

The injury rates established in previous studies (Goldberg et al., 1988;
Goldberg, 1984, Roser and Clawson, 1970; Silverstein, 1979) were in most
cases much lower than this study. There are two main reasons for this; the
definition of a reportable injury and the method of reporting injuries were
different. Even if a study was listed as prospective, there was always some
period of time (usually at least a week) before the injuries were reported. Roser
and Clawson were the only investigators that established a system of daily
reporting of injuries by the coaches. Only injuries that caused a player to miss at
least one week of participation were counted which made the injury rate very low.

In the present study, injury rates were higher among the older athletes

(7th - 8th grades) than younger athletes (4th — 6th grades) and injuries occurred

74

twice as often in games than in practices across the sample population. The
athletic trainer to player ratio was much smaller during games than practices,
which could be a factor in higher injury rates during games. Body size is
considered and advantage in American youth football, however, body size in
general had little impact on the risk of injury in the present study. Only the BMI in
the younger players was a predictor of injury. From the perspective of perception
of risk of injury, concerns related to uncontrollable factors and re-injury appeared
as predictors of injury in the older players. Although estimated biological maturity
status as used in this study was not a significant predictor of injury in youth
football, this study showed the practicality of using predicted adult height as a
maturity indicator rather than traditional methods such as skeletal age.

In summary, results of this study showed that body size, maturity status,
and perception of risk of injury may, in some instances, be predictive of injury in
youth football. However, there are no clear trends throughout the study
population. The greatest differences in injury distribution occurred with younger
players (4-6th grades) who are less mature and have a lower BMI. This is
potentially important information for youth football leagues, coaches, and parents
who are concerned about the size and maturity of the participants. If a small
child or adolescent wants to play football, the risk of injury seems to be greater
for a light and less mature child whereas a lighter and/or less mature adolescent
is at relatively the same risk as a heavier, more mature player. This seems to
contradict the view of many parents who worry about the safety of their child and

urge the child to play another sport, which they feel is safer, or poses less risk of

75

injury. It is difficult to assess whether or not football is any more dangerous than
other sports such as soccer, but knowing that participants, in general, have a
similar risk of injury in youth football is important.

There seems to be a biological influence on the psychology of being
injured in older youth players. This could be the result of the socialization into
football related to body size. Larger, heavier players are more likely to have
positive reinforcement in football than lighter players and, therefore, to have a
lower perception of risk of injury. These results differ from those of Kontos
(2000) in a study of youth soccer players. Heavier players had a higher
perception of risk of injury than lighter players, most likely due to a negative
association between being larger in size in a sport where speed and agility is
preferred. The biological influence seems to be limited to heaviness (BMI) and
not to height or weight. Also, the population of youth football players is taller
and heavier than national reference data, suggesting that direct or indirect
selection of youth football participants. A taller, heavier, and likely more mature
child / adolescent is more likely encouraged to participate in youth football, or the
child himself may want to participate because of his body size.

Results from this study or from similar studies in the future have
implications for youth football and perhaps other youth sports. Injury rates, injury
types and injury severity will be better understood. Hopefully the true risk of
injury in youth football will be made clearer and myths associated with this sport
will be diminished. The significance of body size and biological maturity with

regard to risk of injury and team selection will be better understood and utilized.

76

Injury surveillance and medical coverage at the community level could be
regarded as an important and valuable service. The use of certified athletic
trainers is expanding and there could be a market at the community level.

Future research should focus on comparison of different sports and also
on different leagues of youth football that institute different sets of safety rules.
The Mid-Michigan Pony Football League’s rules are different than Pop Warner
football rules regarding the matching of players. The same type of study on other
sports such as soccer, basketball, ice hockey should be done for comparison of
injury rates as well as examination of biological risk factors. It is possible that
studies will reveal that the incidence and prevalence of injury in other sports are
higher or lower than youth football or that player-related risk factors better
predictors of the risk of injury in those sports. Regardless, there are limited
studies on injuries in youth sports at the local level and a larger focus should be
placed on scientific research at this level because of the sheer numbers of
participants involved.

Longitudinal studies should be conducted to better examine injury
patterns, rates, type and severity. One season of data can give a general
indication of injury rates and patterns. Data collected over a five-year time period
will likely follow the many of the same athletes throughout a career in youth
football (4th-8th grade). Specific to this study, future measurements of the actual
adult height attained by the participants of this study will show the accuracy of

the predicted adult height method used to estimate maturity status.

77

CHAPTER 6
SUMMARY AND CONCLUSIONS

The present study concerned injury in a sample of sports participants 9-14
years of age in a single sport — American youth football — over the course of a
single season. Several specific questions were the focus of the study:

1. What are the prevalence and incidence of injury in youth football?

2. Do injury rates differ grade?

3. Are there biological variables (weight, height, biological maturity) that
are associated with the risk of being injured, and are there interactions among
these variables?

4. Is the perception of risk of injury in youth football related to the
likelihood of injury?

The subjects were 355 youth football players 9-14 years of age in the 4th
to 8th grades from two communities in south central Michigan (Holt [n = 210] and
St. John’s [n = 145]). The heights and weights of the players were measured at
the beginning of the season (August). Boys wore shorts and a t-shirt, and shoes
were removed. The body mass index (BMI) was calculated as weight (kg) divided
by height (meters) squared (kg/m2).

Participants were also asked to complete a 24-item questionnaire dealing
with the perception of risk of injury in sports (Kontos et al., no date). The
questionnaire was administered to the athletes during a practice session. Of the

total sample, 240 (72%) boys returned completed questionnaires.

78

Percentage of predicted adult height was used as an estimate biological
maturity status. The procedure is based on the assumption that the closer an
individual is to his adult height, the greater his level of maturity. The prediction
equation used in the study required the current age, height, and weight of the
boy, and midparent height of his biological parents. Reported heights of both
biological parents were obtained for 296 athletes. Since individuals ordinarily
over estimate self-reported heights, a correction for reported heights was used.
The Khamis-Roche Method for predicting adult height was used. The method
requires the child’s, current decimal age, height and weight, and midparent
height.

Predicted adult heights were estimated for 36 of the 58 subjects for whom
parental data were not available. Mean midparent height of the population was
used in the equation. Adult height was not predicted for the remaining 22 boys
because one or more of the necessary variables for the calculation were not
recorded (age, height or weight). Predicted adult height and percentage of
predicted adult height was estimated for 36 participants for whom parental
heights were not available. The mean midparent height of the population was
used in the prediction equation. Maturity status was estimated by expressing
current height as a percentage of the predicted adult height.

The boys were also divided into maturity categories. An athlete whose
percentage of predicted adult height was > 1 SD above the mean-for-age for the
reference samples (Fels longitudinal study) was considered advanced (early) in

maturity, and an athlete whose percentage of predicted adult height was >1 SD

79

belo
athle

“on I

wee
play
drills

8885

part

Oppl

Org;
durl

DOS

reg
Dra
gar
gar.

talk

below the mean-for-age was considered delayed (late) in maturity status. An
athlete who fell within +/- 1 SD of the mean for age were considered average or
“on time” in maturity.

The leagues established the grouping of participants by grade. The
football season lasted from mid-August until the end of October. The first four
weeks consisted of practice four times per week. It was mandatory for each
player to undergo 8 hours (4 practice sessions) of conditioning before contact
drills could be performed. Each team played six games over the seven-week
season.

A NATA certified athletic trainer (ATC) was on site to record the number of
participants at all practices and games, i.e., coach-directed sessions, which were
opportunities for injury (exposures) and injuries as they occurred. The ATC was
the primary on-site caregiver to injured athletes (as agreed by the league
organizers) and the primary recorder of injury data. If an athlete was injured
during a practice session or a game, the ATC recorded the athlete’s grade,
position or activity at the time of injury, the assessment of the injury (body part,
type of injury), perceived severity, weather and field condition, and the action
taken - return to participation, removal from participation, taken to hospital. With
regard to away games, the ATC consulted with the head coach at the next
practice session to inquire about injuries that may have occurred during the away
game. If the coach indicated that an athlete did sustain an injury at an away
game, the athlete was then interviewed by the ATC. The certified athletic trainer

talked to the parents if the injury was perceived to be serious enough to warrant

8O

being seen by a family or emergency room physician. If a child did seek the
attention of a physician, further contact was made with the parents of the injured
child to attain a more detailed description of the diagnosis. The ATC inquired
about the context of the injury, the body part involved, and any time loss. The
ATC performed an evaluation of the injury and contacted the parents to discuss
possible treatment options. Along with recording injury data, daily logs of
attendance, type of session, activities carried out during the session, and
weather conditions were recorded. This information was used to estimate
exposure and related conditions.

Along with injury information, the ATC recorded exposure data for each
grade. For the purposes of the study, an exposure occurred any time an athlete
participated in a game or practice session and had a chance of injury. League
rules required that a player play at least 6 plays in each half of a game.
Exposure data were recorded for each practice and each game (home and
away). Exposure data allowed for the calculation of injury rates per Athlete
Exposures (AE). Exposure rates were calculated for the entire population, each
community, and each grade, and were also differentiated between practice and
game rates.

Descriptive statistics were calculated for age, height, weight, the BMI, and
percentage of predicted adult height by whole year age group, i.e., 10.0 to 10.99,
11.0-11.99, etc., and by grade. Age-specific means and standard deviations for
height, weight, and the BMI were compared to reference values for American

boys (Centers for Disease Control and Prevention, 2000), and corresponding

81

values for the percentage of predicted adult height were compared to the
corresponding means and standard deviations for boys in the Fels sample.

Chi square (x2) analysis was used to compare actual and expected injury
distributions. Height, weight, BMI, and maturity status were divided into tertiles
as well as a mean split within grade for these analyses. These variables were
also cross-tabulated with injured and non-injured players.

The case control method focused on an injured player (case) and three
randomly selected teammates (controls) who were participants in the
game/practice at which the injury occurred. The cases (injured players) and
controls (non-injured players) were compared.

Body size and maturity status were split into high and low groups using
appropriate methods (i.e., median split, mean split, quartile split) in order to
calculate the odds ratio (OR). Odds ratios were calculated for body size (height,
weight, BMI) and maturity status. The 95% confidence interval was estimated for
each OR. Odds ratios were compared using a Mantel-Haenszel x2 test to
determine the association between each risk factor and injury. Logistic
regression was used to assess their relative and combined contribution to the
prediction of injury in youth football.

Injury rates show an increase with grade, which is logical because players
in higher grades are larger and more force is created with impact and torque. It
is difficult to compare the injury rates of this sample with previous studies
because research is lacking and there are different definitions of an injury in

available research. Although biological variables (height, weight, BMI, maturity

82

status) do not seem to interact to influence the chance of injury, there is some
evidence that individual variables are associated with an increased likelihood of
injury. An increase in BMI is associated with an increased likelihood of injury in
the younger population (4th-6th grade). Of the six factors in the RISSc
questionnaire, only 2 (reinjury and uncontrollable injury) seem to be predictors of
injury, but only in the older population (7th-8th grade).

Youth sports is an area where there are large numbers of participants and
a serious lack of research examining injuries. Most of the few studies that do
exist do not examine player-related risk factors such as height, weight, body
mass index and especially maturity status. These are areas that need
systematic examination due to the variability in body size and maturity of the
participants. After the examination of injuries and body size in a midwestern
youth football league, it is clear that further research needs to be conducted.
Factors that have been reached in this study are:

- Injury rates increase with grade.

0 Most injuries are minor -— not requiring long periods of time before

return to play.

. Most injuries occur to interior linemen.

. Contusions/General trauma are the most prevalent injury type.

- Maturity status expressed as a percentage of predicted adult height

does not seem to influence the chance of injury.

83

Table 1. Injury rates per 1,000 exposures by sex and level of competition of
numerous high school 5 orts over an B-Lear period1.

 

 

 

 

 

 

 

Level BOYS GIRLS
lnterrnediate 0.66 0.69
Junior Varsity 0.91 0.66
Junior Varsity and Varsity 0.63 0.57
Varsity 1.38 0.65
Total 0.89 0.64

 

 

 

1 Adapted from Beachy et al. (1997)

84

 

Table 2. Injury rates among high school football players by source of data
and definition of injury.

 

 

 

 

 

 

 

 

 

Investigator Method of Definition of injury Injury
data rate
collection (%)

Adkison et Daily reports Traumatic medical condition during game 53.03

al., (1974) by a team play resulting in any one of the following:
representative necessitated discontinuation of

participation for the remainder of the game;
resulted in missing two or more practices;
resulted in missing one or more
subsequent games.

Blyth and Weekly, direct Medical treatment or usual activity 48.8

Mueller interview restricted 1 day beyond the injury.

(1974)

Garrick and Weekly, direct One or more practices altered or missed. 81.1

Requa b interview

(1978)

Olson c School Two or more practices missed. 10.5

(1979) generated
reports

Prichett d Insurance Claim filed 24.9

(1980) claims

DeLee and Daily reports Causing to miss all or part of a practice or 50.6

Farney by certified game; treated by a physician; all head

(1992) athletic injuries reported to an athletic trainer.
trainers

Powell and Prospective, Any injury that causes cessation of 37.6

Barber- daily reports participation in the current game or

Foss by certified practice and prevents the player’s return to

(1999) athletic that session;
trainers; direct Any injury that causes cessation of a
interview player’s customary participation on the day

 

 

following the day of onset; Any fracture that
occurs, even though the athlete does not
miss any regularly scheduled session;

Any dental injury, including fillings,
luxations, and fractures; Any mild brain
injury that requires cessation of a player’s
participation for observation before
returning, whether in the current session or
the next session.

 

 

a Games only
b,c,d Adapted from Thompson et al., (1987)

85

 

Table 3. Injury rates among youth football players by source of data and
definition of injury.

 

 

 

 

 

 

Investigator Method of data Definition of Injury rate (%)
collection injury
Roser and Questionnaires filled Missing a 2.3
Clawson out by coaches when a practice or a
(1979) player was injured game
Silverstein Direct observation None 1977 - 18.1
(1979) 1978 — 12.9
Goldberg et Questionnaires Missing at least 15.4
al., (1984) completed by league one practice or
personnel and game session
telephone interview
with parent of injured
child
Goldberg et Retrospective Greater than 5.0
al. (1988) telephone survey at seven days
end of season restriction

 

 

 

 

86

 

Age (years)

4I5th
6th
7th
8th

Height
(cm)

4I5th
6th
7th
8th

Weight
(kg)

4l51h
6th
7th
81h

(kg/m2)

4/5th
6th
7th
8th

Percentage

of

predicted
adult ht.

415th
61h
7th
81h

N

80
41
63
35

76
40
52
34

76
40
52
34

76
40
52
34

73
39
51
33

M

10.1
11.5
12.5
13.5

141.6
149.2
157.0
167.1

42.0
46.7
54.7
66.6

20.6
20.7
22.1
23.7

79.4
83.3
87.2
91.7

Community A

SD

0.67
0.41
0.34
0.36

8.2
6.6
6.5
7.6

12.5
14.3
13.3
15.8

4.3
4.7
4.4
4.6

2.8
2.4
2.5
3.0

Community B

 

N

53
43
42

48
44
41

52
43
45

47
42
41

46
38
37

87

M SD
10.3 .65
11.7 .46
12.8 .47
143.0 7.1
151.9 7.0
160.3 8.2
41.4 11.5
50.1 12.0
57.2 13.9
19.9 4.1
21.5 3.9
22.2 4.3
79.9 2.7
84.2 2.3
88.7 2.6

Table 4. Characteristics of players by grade and community.

N

1 33
84
95
35

1 24
84
93
34

128
83
97
34

123
82
93
34

1 19
77
88
33

12t_a.|
M

10.2
11.6
12.6
13.5

142.1
150.6
158.5
167.1

41.8
48.5
55.9
66.6

20.3
21.1
22.1
23.7

79.6
83.7
87.8
91.7

SD

0.66
0.46
0.42
0.36

7.8
6.9
7.4
7.6

12.1
13.2
13.6
15.8

4.2
4.3
4.4
4.6

2.7
2.4
2.6
3.0

Table 5. Mean corrected parental heights and midparent heights (cm).

Mean SD Minimum Maximum
Father 178.4 6.6 153.6 202.4
Mother 164.9 6.3 151.2 183.8
Midparent 171.7 4.8 153.1 191.9

88

 

Table 6. First organized sports reported by the study population.

Sport N %

Tee ball 103 35.5
Soccer 90 31.0
Baseball 19 6.6
Football 19 6.6
Floor Hockey 16 5.5
Basketball 13 4.5
Other * 9 3.1
Ice Hockey 7 2.4
Wrestling 6 2.1
Softball 5 1.7
Swimming 3 1.0

* Other — martial arts, bowling, track & field

89

 

Table 7. Prevalence of injured players during the season by grade and age
group.

Grade
4-5th
6th
7th
81h

Total

Age

10
11
12

13l14

Total

134

87

99

35

355

30

54

87

81

82

347

N Injured
24
19
39
18

100

N Injured
5
8
19
22

35

98

90

%

17.9
23.0
39.4
51.4

28.5

°/o

16.7
14.8
21.8
27.2

42.7

28.2

Table 8. Types of reportable injuries sustained during the football season,
communities combined.

Injury Frequency %

General Trauma 55 40.0
Strains 20 14.6
Sprains 18 13.1
Others 1 13 9.5
Neurotrauma 12 8.8
Fractures 10 7.3
Overuse 4 2.9
Non-specified 5 2.2

1 Others include upset stomach, general feeling of malaise, headache, and other
illnesses that could have been heat related. However, there was only one case
of heat exhaustion.

91

Tal

SEE

Po:

Off

De'

Ru

Drl

De

W i

Table 9. Frequency of reportable injuries by position or activity during the
season, communities combined.1

Position/Activity Frequency %
Offensive Line 25 18.2
Defensive Line 25 18.2
Running Back 22 16.1
Drills 15 10.9
Defensive Back 14 10.2
Wide Receiver 8 5.8
Quarterback 8 5.8
Linebacker 6 4.4
Special Teams 5 3.6
Non-specified 7 6.6

1 Position refers to the specific position that an athlete was playing when the
injury occurred. Many boys played more than one position.

92

Table 10. Athlete exposures by practices, games, and both combined by
community.

Practices

Community A Community B Combined
Communities
4/5th grade 2143 1486 3629
6th grade 1150 1265 2415
7th grade 1567 1320 2887
8th grade 1156 0 1156
Games
4/5th grade 453 306 759
6th grade 240 260 500
7th grade 383 288 671
8th grade 291 0 291
Practice 8: Games
Combined
4/5th grade 2593 1792 4388
6th grade 1390 1525 2915
7th grade 1950 1608 3558
8th grade 1447 0 1447
Total 7383 4925 12308

93

Table 11. Estimated practice rates per 1000 athlete exposures (AE) by

grade and community.

4I5th

6th

7th

8th

Total

Community A

5.6

11.3

12.8

14.7

10.6

Community B
6.7
2.3

15.9

8.1

Table 12. Estimated game rates per 1000 athlete exposures (AE) by grade
and community.

4I5th
61h
7th
8th

Total

Community A

11.0

25.0

21.1

20.6

18.3

Community B
13.1
7.7

31.3

17.6

94

Table 13. Estimated game and practice rates combined per 1000 athlete
exposures (AE) by grade and community.

Community A Community B
4I5th 6.5 7.8
6th 13.7 3.3
7th 14.2 18.7
8th 15.9
Total 12.1 9.7

Table 14. Estimated game and practice injury rates, and rates for games
and practices combined per 1000 athlete exposure (AE) by grade in the
total sample.

Practices Games Total
4/5th 6.1 11.9 7.1
6th 6.6 16.0 8.2
7th 14.2 25.8 16.3
8th 14.7 20.6 15.9
Total 9.6 18.0 11.1

95

Table 15. Estimated injury case rates and player rates per 100 players by
community, and communities combined.

Injury Rate
Case Rate

Player Rate

Communities Communities
A B Combined
42.4 33.1 38.6

30.5 24.8 28.4

Table 16. Estimated player rates and case rates per 100 players by grade.

Grade

4/Sth

6th

7th

8th

Player Rates Case Rates
18 24
22 26
39 60
51 66

96

Table 17. Odds ratios of being injured for upper (above the mean) and
lower (below the mean) values of height, weight and body mass index.

4I5 grades
Variable Low High
Height 1.7 0.6
Weight 2.5 0.4
BMI 3.2 0.3
6th grade
Height 1.0 1.0
Weight 1.5 0.7
BMI 1.8 0.6
7th grade
Height 0.7 1.5
Weight 0.7 1.5
BMI 0.7 1.5
8th grade
Height 1.6 0.6
Weight 1.6 0.6
Body mass index 1.0 1.0

97

Table 18. Reliability of the RISSc questionnaire.

Factor YOUNGER - OLDER -
Alpha Alpha
Uncontrollable .75 .77
Controllable .80 .78
Overuse .75 .74
Upper Body .75 .74
Surface .64 .63
Related

Re-Injury .60 .62

98

Table 19. Logistic Regression final model in 4th-6th grade.

95% Confidence Interval for EXP (B)
Variables EXP (B) Lower Upper

BMI .426 .210 .867

Table 20. Logistic Regression final model in 7th-8th grade.

95% Confidence Interval for EXP (B)

Variables EXP (B) Lower Upper
Uncontrollable .438 .099 .965
Reinjury 3.883 1.629 9.256

 

Table 21. Partial correlations for biological variables (height, weight, BMI,
maturity) and the Uncontrollable RISSc factor in the older population.

 

 

 

 

 

 

Correlation Significance
Height -.2662 .009
BMI -.1873 .071
Weight -.2516 .014
Maturity -.2029 .052

 

 

 

 

99

170 h

 

 

 

 

 

 

 

 

 

 

 

E 160 /: ——CDC 50th
2 150 !/';/ percentile
E140 (f/ *Sfrieliit.
a:
130 - 0- 'Youth
120 e 4 . . , , Football
9 1o 11 12 13 P'aye's
Age, years

Figure 1. Mean heights of youth football players relative to reference
values for United States boys.

100

N
O

 

 

+ CDC 50th
percentile

O)
O
\

 

 

01
c
Q
\

—:—<—- CDC 75th
percentile

b
O

 

Weight, kg
Q
\

 

 

. .

~ . - 0- 'Youth l
Football "

Players

(A
O

 

 

 

 

N
O
.l

 

910111213 I;

Age, years

 

Figure 2. Mean weights of youth football players relative to reference
values for United States boys.

101

 

—o— CDC 50th
percentile

-—:-ee— CDC 75th
percentile

- 0- 'Youth
Football
Players

 

 

 

 

9 10 11 12 13

Age, years

Figure 3. Mean BMl’s of youth football players relative to reference values
for United States boys.

102

 

 

CD
01

\

 

 

on
01

 

 

-O—Fels
- 0- 'Football

 

 

oo
o
l

\

 

 

‘1
0'1

 

 

% Predicted Adult Height

 

 

N
o
q
.4

9 10 11 12 13 14

Age, years

 

Figure 4. Mean percentage of predicted adult height in youth football
players compared to corresponding data for males from the Fels Research
Institute longitudinal growth study upon whom the height prediction
equations were based

103

Appendix I

 

MICHIGAN STATE
u N 1 v E R s I T Y
June5,2001

To Robert MALINA
129 IM Sports crcie

RE“. IRB ’ 00-282 CATEGORY: FULL REVIEW
RENEWAL APPROVAL DATE: June 4. 2001

TITLE: INCIDENCE AND PLAYER RISK FACTORS FOR INJURY IN YOUTH FOOTBALL

The University Committee on Research Involv'ng Human Subjects' (UCRIHS) review of this project
iscompleteandIampleasedbadvisematttnringmdwefluedmeWswjedsappeub
be adequately protected and methods to obtain Informed consent are appropriate. Therefore, the
UCRIHS APPROVED THIS PROJECTS RENEWAL

Thls letter also notes approval for addition of Co-Pls (Morano and Barron), revised
consent form.

RENEWALS: UCRIHS approval is valid for one calendar year, beginning with the approval date
shownabove. ProjectsconUnuhgbeymdmeyeunmtberenewedwlmmegreenrenewelm
A maximum of tour such expedited renewal ere possble. Investigators wishing to continue a project
beyond Ihattimeneedtcsubmitltagainfor completereview.

REVISIONS: UCRIHS must review any changes in procedures Involving human subjects, prior to
hitiationofthechange. llmtsbdmeatthetiriedreriewal.pleaseusemeoreenrenewalfumTo
revise anapproved probodetenyobterthneduhgheyeaxsendywvwittenrequesttom
UCRIHS Chair, requesting revised wont and referencing the project's IRS:- and title. W. In
your request a desorption of the change and any revised hstrurnents, consent turns or
advertisements that are applicable.

PROBLEMS/CHANGES: Should either of the following arise durlng the course of the work notify
UCRIHS promptly: 1) problems (mexpecled side effects, complaints, etc.) lnvoMng human subjects
cr2)changeslnlheresearch environMornuhIonnannhdicathgoreaternsktomemnm
subjectsthan existedMienmeprotocolvnsprei/stly reviewed andepproved.

 

If we can be of further assistance, please contact us at 517 355-2180 or via email:
0““ 1' UCRIHS@prlot.msu.edu. ’
RESEARCH

”10

Sincer ,
amount /
STUDIES 1W
mm ir Kamar, MD.

Mae 3‘1.“ Interim Chair. UCRIHS

w Star on»,
241mm
when. Hideout
ems-106
517059210
wt snmnsrs AK: br

no mm mm

(“mm cc: Peter Morano

128 IM Sports Circle

Dew-Seem
magnum
hutch/da-
ears-w
M”

104

Appendix II

Participant Consent Form

Michigan State University
Institute for the Study of Youth Sports
213 IM Sports Circle
East Lansing, MI 48824-1049
517 355-7620

Thank you for participating in this study designed to assess the thoughts you
have concerning being injured when playing sports. This study will provide
information concerning the events that might lead to injury in youth sport
participants.

For this study, you will be asked to complete a questionnaire regarding your
thoughts on being injured in sports. We will also measure your weight and
height. This will be done at the beginning of the season at equipment handout
and will take approximately 10 minutes of your time. In addition, any injury or
injuries that you have during your football season will also be recorded by a
certified athletic trainer. If you are injured, we will call you to discuss your injury in
more detail. You will also be asked to provide written information about your age,

previous injuries, and experience in sport.

All data that you provide, and the results of this study will be confidential and
anonymously reported. You will be assigned a coded identification number to be
used on all information. All questionnaires and individual data will be stored in a
locked area accessible only to the investigators of the study. Only group data will
be used in any reporting or future use of the information from this study. Group
results will be made available to you on request.

Participation in this study is voluntary. You may choose not to participate at all,
refuse to answer certain questions, or withdraw from the study at anytime,
without penalty.

Any questions concerning participation in this study should be directed to Robert
M- Malina, Professor of Kinesiology, 517 355-7620. If you have additional
Questions or concerns about your rights in this research study, please feel free to
contact David Wright, Michigan State University’s Chair of the Committee on
Research Involving Human Subjects at 517 355-2180.

105

:L-. t -:-d‘v'u‘-'i-.'a‘l.'.-.t

 

Thank you for your time and cooperation.

I have read the above description of this study. I understand my rights as a
participant and agree to participate in this study.

Please Print:

 

First Name Initial Last Name

 

 

Signature Date

 

106

Appendix III

Parental Informed Consent Form

Michigan State University
Institute for the Study of Youth Sports
213 IM Sports Circle
East Lansing, MI 48824-1049
517 355-7620

Dear Parents/Guardians:

Hello! Robert M. Malina, Ph.D., Professor of Kinesiology and Mary Barron (ATC,
certified athletic trainer), a master’ student and another master’s student (an
ATC, to be named) at Michigan State University are currently working on a study
entitled “Incidence and Player Risk Factors for Injury in Youth Football.” Dr.
Jeffrey Kovan, Head Physician for the MSU Sportsmedicine Team, is a
consultant for the project. This study will assess the relationship between the
perception of risk of injury and body size on injury occurrence among youth
football participants.

The study will involve your son’s participation at the beginning and end the
football season in completing a questionnaire designed to learn more about his
thoughts regarding injuries in sport. Your son’s height and weight will also be
measured at the beginning of the season. You will be asked to provide
information on your son’s previous experiences in youth sports and about injuries
in sport if he has had any. In addition, we will ask you to report your own height,
that is, that of the boy’s father and mother. This information will be used to
estimate how tall your son will be as a young adult, so that we can express his
current height as a percentage of his predicted adult height. This will provide us
with an estimate of his maturity status, boys closer to their predicted adult height
are more mature than boys who are further removed from their adult height.

During the course of the football season, the certified athletic trainer participating
in the study, will record information concerning injuries that occur during a
practice session or a game during the course of the season. If your son is
injured, we will contact you by phone to obtain more detailed information about
the injury (type, severity, treatment). With your permission, we also will discuss
the injury with your son to get his view of what happened and how it might
influence his return to the sport.

107

 

All identities and recorded information from the study will remain confidential and
be analyzed using individual identification numbers. Participants will remain
anonymous in any reporting of the data from the study.

In order for us to complete the study, we will need your written consent in the
space below to allow your son to participate in the study. Participation in the
study is voluntary and your son can decide to discontinue participation at any
time. If your child decides to discontinue participation, his will not be used.

Any questions concerning participation of your child in this study should be
directed to Robert M. Malina, Professor of Kinesiology, 517 355-7620. If you
have additional questions or concerns about your son’s rights in this research
study, please feel free to contact David Wright, Michigan State University’s Chair
of the Committee on Research Involving Human Subjects at 517 355-2180.

I, agree to allow my child

 

Your name-printed Your child’s name-
printed

to participate in this study.

 

 

Your signature Date

Thank you for your consent for your child’s participation in this study. Please
have your child return the completed form to us at the equipment handout
session.

Thank you,

Robert M. Malina, Ph.D., FACSM
Department of Kinesiology

128 IM Sports Circle

Michigan State University

East Lansing, MI 48824

517 355 7620
RMALINA@piIot.msu.edu

108

 

Appendix IV

Background in Sport Information

Child’s Name
Date of Birth:
Today's Date:

 

 

 

How old was your child when he first began to play on an organized sport team
(such as soccer, t-bail, football) that practiced and played a regular schedule of
games, or when he first began to regularly practice and compete in an organized
sport program (such as football, swimming)? Organized means that you had an

assigned coach for your team or sport.

 

What was the first organized sport that your son played?

 

What other organized sports has your son played and how many years has he

played each sport?

SPORT YEARS

 

 

 

 

 

109

 

Has you son ever been injured during a sport practice or during a
game/competition?
Yes No

Please continue on the next page
If yes, please circle the appropriate answer:

1. What specific body part was injured?

 

head/neck face shoulder/arm forearm-wrist-hand
trunk hip-thigh-Ieg knee ankle-foot other

2. What type of injury was it? if"
sprain/strain fracture laceration general trauma

3. Did your son receive treatment? Yes No If yes, was he

treated at:

an emergency room or hospital Yes No

a doctor’s office Yes No

at home Yes No
4. Did your son miss any games/competitions/practices due to the injury? Yes

No
In evaluating the height and weight of your son, it is important to know the size of
the biological parents. Please report the height of both parents to the nearest 1/4
inch (without shoes):
Father’s height

Mother’s height

110

 

Appendix V

Game/Practice Information Form

DATE

 

LOCATION

 

TYPE OF SESSION: PRACTICE

 

NUMBER OF ATHLETES

 

LENGTH OF SESSION min.

IF PRACTICE: SPECIFIC ACTIVITIES

GAME

 

 

 

 

 

 

 

 

111

 

 

Appendix VI

Injury Report Form

NAME DATE

 

Athletic Session

 

 

 

 

Game: Warm-up 1St Quarter 2"d Quarter 3rd Quarter 4th
Quarter I
Practice:
Position of injured played: Offense Defense
Type of surface Natural Artificial 1:
Surface condition Dry Wet Muddy
Frozen
Weather conditions:,Hot_Warm_Cool Cold Rain Snow

Point in Season

 

Action Taken: Removed from participation and returned immediately
Returned from participation and returned after resting
Removed from remainder of participation
Taken to hospital by parent

Taken to hospital by ambulance

 

 

 

112

 

Clinical Impression:
Injured Part of Body:

Head Neck Shoulder UpperArm Elbow/Forearm HandNVrist/Fingersfl’humb
Hip Thigh Knee Shin Calf Ankle Foot/Toe(s)
Back Abdomen Chest Other

Type of Injury:

Sprain Strain Fracture General Trauma Neurotrauma
Laceration Overuse Other
Perceived severity of injury: Mild Moderate Severe !

Summary of Evaluation:

 

 

 

 

113

Appendix VII

Risk of Injury in Sports Scale (RISSc)

Please indicate how likely you think it is that the following events will happen to you
while playing your sport.

WHAT DO YOU THINK ARE THE CHANCES THAT YOU WILL (Circle your
answers):

Very Some Some Very
Unlike Unlik what what Like Like
ly ely unlikel likely 1y 1y

Y

1. Injure yourself in a collision 1 2 3 4 5 6
with an opponent?

2. Have the same injury that 1 2 3 4 5 6
someone else on your team
recently had?

3. Re-injure an area that you have 1 2 3 4 5 6
recently injured?

4. Be injured in a practice? 1 2 3 4 5 6

5. Fall down and injure yourself? 1 2 3 4 5 6

6. Be injured from a foul or ‘cheap 1 2 3 4 5 6
shot’ by an opponent?

7. Be injured by more aggressive 1 2 3 4 5 6

opponents?

114

 

10.

ll.

12.

l3.

14.

15.

16.

17.

Be injured running into an
object on the field or court (e.g.,

goal posts, vault, boards, etc.)?

Be injured by bigger or stronger

opponents?

Be injured from not ‘taking a

break’ from your sport?

Be injured trying to perform a

skill that you have just learned?
lnj ure yourself on a poor
playing surface (e. g., wet or

bumpy field, dusty floor, etc.)?

Be injured from playing too

many sports at the same time?

Be injured performing a skill

that is hard for you to do?

Injure your ankle?

Be injured practicing too hard?

Be injured by not paying

attention to what you are doing?

H

115

 

18. Inj ure your neck or spine 1 2 3 4 5 6

19. Be injured from competing too 1 2 3 4 5 6
hmd?

20. Be injured by losing your focus 1 2 3 4 5 6
while playing your sport?

21. Trip and injure yourself? 1 2 3 4 5 6

22. Injure yourself on a dangerous 1 2 3 4 5 6

 

piece of equipment?

23. Injure your arm or wrist? l 2 3 4 5 6

24. Injure your shoulder? l 2 3 4 5 6

Factors and Corresponding Items:
‘Uncontrollable’- 1, 6, 7, 9; ‘Controllable’- 8, ll, 14, 17, 20, 22; ‘Overuse’- 10, 13,
l6,19;‘Upper-body’- 18, 23, 24; ‘Surface-related’- 5, 12, 15, 21; ‘Re-injury’- 2, 3, 4.

 

116

flBLIOGRAPHY

Adkison JW, Requa RK, Garrick JG. (1974). Injury rates in high school football.
Clin Orth Rel Res 99: 131-136.

Backous DD, Friedl KE, Smith NJ, Parr TJ, Carpine WD. (1988). Soccer injuries
and their relation to physical maturity. Am J Dis Children 142: 839-842.

Backous DD, Farrow JA, Friedl KE. (1990). Assessment of pubertal maturity in
boys, using height and grip strength. J Adol Health Care 11: 497-500.

Backx FJG, Erich WBM, Kemper ABA, Verbeek ALM. (1989). Sports injuries in
school-aged children: An epidemiologic study. Am J Sports Med 17: 234-240.

Bayer LM & Bayley N. (1959). Growth Diagnoses. Berkley, CA: University of
California Press.

Beachy G, Akau CK, Martinson M, Olderr TF. (1997). High school sports injuries
— a longitudinal study at Punahou School: 1988 to 1996. Am J Sports Med 25:
675-681.

Bijur PE, Trumble A, Harel Y Overpeck MD, Jones D, Scheidt PC. (1995). Sports
and recreation injuries in US children and adolescents. Arch Ped Adolesc
Med 149: 1009-1016.

Blyth CS, Mueller F0. (1974). Football injury survey. Part I: When and where
players get hurt. Phys Sportsmed 2:45-52 (Sept).

Brozek J, Grande F, Anderson JT, Keys A. (1963). Densitometric analysis of
body composition: revision of some quantitative assumptions. Ann N Y Acad
Sci; 110: 113-140.

Carnegie Corporation of New York (1996) The Role of Sports in Youth
Development. Report of a meeting convened by the Carnegie Corporation.
New York: Carnegie Corporation of New York, pp 1-157.

Centers for Disease Control and Prevention. (2000). National Center for Health
Statistics CDC growth charts: United States.
_h_ttg//www.cdc.qov/growthchafls.htm.

Culpepper MI, Neimann MW. (1983). High school football injuries in Birmingham,
Alabama. So Med J 76: 873-878.

Dick Butkus Football Network. (2001).

117

 

http://www.dickbutkus.com/dbfn/popwarner/index.html
DeHaven KE, Lintner DM. (1986). Athletic injuries by age, sport, and gender. Am
J Sports Med 14: 218-224.

DeLee JC, Farney WC. (1992). Incidence of injury in Texas high school football.
Am J Sports Med 20: 575-580.

Ellison LF, Mackenzie SG (1993) Sports injuries in the database of the Canadian
Hospitals Injury Reporting and Prevention Program - an overview. Chron Dis
Canada 14:96-104.

Epstein LH, Valoski AM, Kalarchian MA, McCurley J. (1995). Do children lose
and maintain weight easier than adults: A comparison of child and parent
weight changes from six months to ten years. Obes Res 3: 411-417.

 

Godshall RW. (1975). Junior league football: Risks vs benefit. Sports Medicine
3:139-144.

Goldberg B, Rosenthal PP, Nicholas JA. (1984). Injuries in youth football. Phys
Sportsmed 8: 122-130 (Dec).

Goldberg B, Rosenthal PP, Robertson LS, Nicholas JA. (1988). Injuries in youth
football. Pediatrics 8: 255-261.

Jackson AS, Pollock ML. (1985). Practical assessment of body composition.
Phys Sportsmed; 13: 76-89.

Khamis HJ, Roche AF. (1994). Predicting adult stature without using skeletal
age: The Khamis-Roche method. Pediatrics 94: 504-507.

Kontos A. (2000). The effects of perceived risk, risk-taking behaviors, and body
size on injury in youth sport. Doctoral dissertation, Michigan State University,
East Lansing, MI.

Kreipe KE, Gewanter HL. (1985). Physical maturity screening for participation in
sports. Pediatrics 75: 1076-1080.

Kvist M, Kujala UM, Heinonen OJ, Vuori IV, Aho AJ, Pajulo O, Hintsa A, Parvinen
T. (1989). Sports-related injuries in children. Int J Sports Med 10: 81-86.

Malina RM. (1995). Anthropometry. In PG Maud, C Foster, eds, Physiological
Assessment of Human Fitness. Champaign, IL: Human Kinetics, p 205.

Malina RM. (2001). Injuries in organized sports for children and adolescents (ed):
Children and Injuries. J. Frost, Lawyers and Judges, Tucson, AZ.

118

Malina RM, Beunen G. (1996). Matching of opponents in youth sports. In 0 Bar-
Or (ed):(lV): The Child and Adolescent Athlete. Champaign, IL: Human
Kinetics, pp 202-213.

Malina RM, Bouchard C. (1991). Growth, Maturation, and Physical Activity.
Champaign, IL: Human Kinetics.

Malina RM, Katzmarzyk PT. (1999). Validity of the body mass index as an
indicator of the risk and presence of overweight in adolescents. Am J Clin
Nutr 70 (suppl): 1318-1358.

MacMahon B, Trichopoulos D. (1996). Epidemiology: Principles and Methods.
Boston, MA: Little, Brown and Company.

Mid-Michigan Pony Football League. (2000). Rules

Mueller F, Blyth C. (1982). Epidemiology of sports injuries in children. Clin
Sports Med 1: 343-351 (Nov).

NEISS Data Highlights. (1981). National Electronic Injury Surveillance System, 5:
Oct- Dec.

Petersmarck K. (1988). The prevention of injuries in amateur football. Lansing:
Michigan Department of Community Health, DCH-046.

Powell JW, Barber-Foss KD. (1999). Injury patterns in selected high school
sports: A review of the 1995-1997 seasons. J Ath Tr 34: 277-284.

Roche AF, Tyleshevski F, Rogers E. (1983). Non-invasive measurement of
physical maturity in children. Res Quart Ex Sport 54: 364-371.

Roser LA, Clawson DK. (1970). Football injuries in the very young athlete. Clin
Orthop 69: 219-223.

Silverstein BM. (1979). Injuries in youth football. Phys Sportsmed 9: 105-111
(Sept).

Simple Interactive Statistical Analysis. (2001).
htth/homeclara.net/sisaltwthIg.htm#Odds

Smith AD, Andrish JT, Michell LJ. (1993). The prevention of sports injuries of
children and adolescents. American College of Sports Medicine Current
Comment. Indianapolis, IN: American College of Sports Medicine.

Tanner JM. (1962). Growth at Adolescence, 2nd Edition. Oxford: Blackwell
Scientific Publication.

119

 

Thompson N, Halpern B, Curl WW, Andrews JR, Hunter SC, McLeod WD.
(1987). High school football injuries: Evaluation. Am J Sports Med 15: 117-
124.

Tursz A, Crost M. (1986). Sports-related injuries in children. Am J Sports Med 14:
294-299.

Violette RW. (1976). An epidemiologic investigation of junior high school football
injury and its relationship to certain physical and maturational characteristics
of the players. Doctoral dissertation, University of North Carolina, Chapel Hill.

Watkins J, Peabody P. (1996). Sports injuries in children and adolescents treated
at a sports injury clinic. J Sports Med Phys Fit 36: 42-48.

Zaricznyj B, Shattuck LJM, Mast TA, Robertson RV, D’Elia G. (1980). Sports-
related injuries in school-aged children. Am J Sports Med 8: 318-323.

120

 

Iljlijjjlljjjjlljljl