FBepidem

From: Epidemiology of Sports Injuries, D.Caine, C. Caine, K. Lindner (eds.), Champaign, IL: Human Kinetics, pp. 41-62 (1996).

Chapter 4

American Football

Frederick Mueller, Ph.D. University of North Carolina, Chapel Hill

Eric D. Zemper, Ph.D. Exercise Research Associates of Oregon, Eugene

Arlene Peters, M.S. University of North Carolina, Chapel Hill

1. Introduction

Over the past 20 years, football, like a number of other sports, has shown a significant increase in the number of participants. In 1992 there were 1,500,000 high school and junior high school players and 75,000 college and university athletes (68, 69). Since 1971 the number of high school participants has increased by 25% (300,000 players), while participation at the college level has remained steady at approximately 75,000 players. With this increased number of athletes playing football, we would expect there to be an increase in the total number of injuries each year, and an increased number of football injuries has been documented (57). Of course, this translates into an increased burden on the health-care system. Knowledge of the variables influencing and contributing to the incidence of football-related injuries is important to all those involved in the game. Such information allows for the comprehensive detection and prevention of football-related injuries, making the game safer and more enjoyable for everyone.

To better educate players, coaches, athletic trainers, physicians, and sport administrators about the likelihood of injury, it is necessary to identify the causal and influential factors and mechanisms contributing to the risk of injury in the game of football. The purpose of this chapter is to document and describe football injury research studies of players at the high school and college levels. In comparing, contrasting, and evaluating this injury research literature, the goal is to give a detailed picture of the injury situation in football.

We undertook a search of the available research literature on football injuries since 1971, utilizing our personal resources as well as searches of Medline and Sport Discus and using football (injury or accident) and language-English for the on-line search request. Football is one of the few sports that does have a sizeable body of published literature on injuries, and we retrieved a large number of titles. However, relatively few published studies can be considered true epidemiologic studies, prospective or even retrospective. Most of the studies in the published literature are reports of one or two cases of an unusual nature or a case series of football injuries seen in a particular hospital or clinic. These types of studies provide no information on the population from which the injured athletes were drawn, and therefore can provide no information on football injury rates. With a few exceptions, we did not include such case reports or case series in the literature reviewed in this chapter in order to make the reviewing task more manageable and to keep the focus on true epidemiologic data. For similar reasons, we did not include studies involving data collection methods such as review of insurance claims because they provide no data on the population at risk, and it should be obvious that not all injuries result in insurance claims.

While reviewing these studies, it became clear that injury rates were calculated differently across different studies and the definition of a reportable injury varied widely. Any attempt to compare across studies, therefore, was very difficult and, in most cases, meaningless unless they happened to be using similar rates and definitions. Those reporting high school data tended to report their injury distributions as a percentage of the total number of injuries recorded or as the number of injuries per 100 players. At the collegiate level most researchers reported their results as injuries per 1,000 athlete-exposures (A-E), in which one A-E is one athlete participating in one game or practice where he is exposed to the possibility of being injured.

Although the rate of injuries per 100 players is frequently used in the literature, it does not take into account the varying number of games and practices across teams, or that all players are not involved in every practice or game. This can result in some very misleading comparisons, even if two studies report rates per 100 players (104, 108). Reporting injury rates per 1,000 A-E compensates for these differences and allows for more direct comparisons between studies, assuming the studies used similar definitions of a reportable injury. These problems in making comparisons across studies should be kept foremost in the reader's mind while considering the results from the various studies summarized in this chapter.

2. Incidence of Injury

2.1 Injury Rates

The relatively high number of injuries seen in football is the result of several factors. First, the dramatic increase in participation has produced an increase in the total number of injuries (57). Second, the inherently violent nature of the game, the physically demanding aspects of the game, and the speed, strength, and size of the players combine to make football a high risk sport (102).

The total injury rates found in various studies of high school and college football teams are presented in Tables 4.1 and 4.2, respectively. From these two tables a few observations can be drawn:

· The rate of injuries at the high school level as a percent of the players on the team (essentially an injury rate per 100 players), which was most often reported at the high school level, ranged from 11.8 to 81.1. For the reasons mentioned previously, trying to make comparisons across these studies using this type of "rate" carries very little validity. The wide range of results probably stems largely from the variety of definitions of a reportable injury used in the various studies and whether medically trained individuals were recording the data. Among studies with similar injury definitions, the range was 48.8 to 81.1.

· The one study that reported an injury rate per 1,000 A-E (4) reported a rate of 8.1 injuries per 1,000 A-E, which was similar to the values reported in Table 4.2 for college teams. Because this study reported the same type of rate and used the same definition of a reportable injury, a direct comparison with the college studies is possible.

· In the college level studies (Table 4.2) the injury rates per 1,000 A-E all fall within the range of 6.1 to 11.1 (for those studies that used the same injury definition of missing one day or more). The older studies tended to have the higher rates and the more recent studies the lower rates, indicating a possible trend toward slightly decreasing injury rates over the past 15 years.

· It is apparent from a comparison of the results in Tables 4.1 and 4.2 that when a common injury rate is reported and a common injury definition is used, as most studies at the college level did (Table 4.2), direct comparisons between studies are much easier. This is contrasted with the situation in Table 4.1 in which the high school reports tended to use a less rigorous type of injury rate and a wide variety of definitions of a reportable injury.

_______________________________________________________________________________________

Table 4.1 A Comparison of Injury Rates in High School Football

_______________________________________________________________________________________

Study	Type	Duration	Reportable Injury Definition	# subjects/ # teams	# injuries	Inj. per 100 players	Inj. per 1000 A-E
Alles et al., 1979	P ^a	1975-77	Time loss ³ 1 day	2764^b/53	1696^b	63.4^b	8.1
Blyth & Mueller, 1974	P	1969-72	Time loss ³ 1 day	8776/43	4287	48.8
Garrick & Requa, 1978	P	1973-75	Time loss ³ 1 day	624/8	506	81.1
Hoffman & Lyman, 1988 (game only)	P	1986	Prevented from continuing play	1180/14	139	11.8
NATA, 1987	P	1986	Time loss ³ 1 day	6544/105	4292	65.6^b	8.2^b
Olson, 1979	P	1970-78	Time loss ³ 2 days or 1 game	3300/?	465	14.1
Prager et al., 1989	P	1982-85	Time loss ³ 2 days	598/4	251	42.1

_______________________________________________________________________________________
^{a P = prospective cohort study.

^{b These figures were not presented in the original article, but were calculated based on data provided in the article.

_______________________________________________________________________________________

_______________________________________________________________________________________
Table 4.2 A Comparison of Injury Rates in College Football
_______________________________________________________________________________________

Study

Type

Duration

Reportable Injury Definition

# subjects/ # teams

# injuries

Inj. per 100 players

Inj. per 1000 A-E

Alles et al., 1979

P ^a

1975-77

Time loss ³ 1 day

13416/148^b

12432^b

92.7^b

11.1

Buckley&Powell, 1982

P

1975-80

Time loss ³ 7 days

28419/309

7190^b

25.3

3.1

Canale et al., 1981

P

1975-70

Time loss ³ 1 day

265/1

283

46.6

Clarke & Miller, 1977

P

1975-76

Time loss ³ 7 days

?/38

7.1

NCAA, 1990

P

1984-89

Time loss ³ 1 day

?/408

19243

6.4

Whiteside et al., 1985

P

1972-83

Unable to function in usual capacity

2730/1

2186

80.1^b

Zemper, 1989b

P

1986-88

Time loss ³ 1 day

8325/80

3744

45.0

6.1

_______________________________________________________________________________________
^{a P = prospective cohort study.

^{b These figures were not presented in the original article, but were calculated based on data provided in the article.

_______________________________________________________________________________________

2.2 Reinjury Rates
The rate of reinjury to football players appears to be a neglected area of research in the literature. This area warrants further research because it appears that prior injury may be a risk factor for further injury. What little information is available comes from three of the studies reviewed here. In a 4-year study Blyth and Mueller (10) found that 60% of high school players with a prior injury had another injury, whereas only 40% of players without any prior injury had a new injury The NATA study (64) documented that over 10% of the injuries reported in the 1986 study of high school players were reinjuries to a previously hurt area and that 4% of those injuries were directly related to the prior injuries. Although this study indicated a rate of 65 injuries per 100 players, it was noted that only 37% of the players in the study were injured at least once, implying that a considerable proportion of the players were injured more than once, though not necessarily to the same body part. At the collegiate level, Zemper (104, 105) noted that although there was an injury rate of 45 per 100 players, only 33.2 players out of 100 incurred at least one injury, implying that the remaining injuries were recurrent injuries or new injuries to a different body part in a player who had been injured previously.

2.3 Injury in Practice and Competition
The authors of many studies discuss whether practices or games are more dangerous or carry a higher risk. However, most studies make comparisons based on a percentage breakdown of injuries occurring in practices and games, and conclude that because more injuries (or a greater percentage) occur in practices, they are riskier for a player than games. For instance, the 1986 NATA high school study (64) states that 62% of the recorded injuries occurred in practices and only 38% in games. When considering this type of data, it must be remembered that a percentage breakdown of total injuries is not a rate and therefore has no relation to estimating risk. Because a team has at least five or six times as many practices as games, and usually more players participate in practice than games, it would be expected that the total number (or percentage) of injuries would be higher in practices. This type of information has its importance; as the NATA summaries correctly point out, more injuries occur in practices in high school, where there is less likely to be adequate medical coverage. This is a situation that needs to be addressed. But this does not mean that an individual player is at a greater risk of injury in practices. Among the few studies that correctly address this issue are the collegiate studies done by the NCAA (67) and by Zemper (104, 105), in which injury rates are calculated separately for practices and for games, and a relative risk can therefore be calculated based on these rates. The NCAA study (67) provides a practice injury rate of 3.99 per 1,000 A-E and a game injury rate of 35.45 per 1,000 A-E, from which a relative risk of 8.9 can be calculated. This indicates that an individual player is 8.9 times as likely to be injured in a game as he is in a practice, even though the NCAA data also show that the majority (57%) of the injuries occur in practice. The data from Zemper (105) show a practice rate of 3.79 per 1,000 A-E and a game rate of 32.11 per 1,000 A-E, so the game injury rate in this study is 8.5 times higher than the practice rate.

3. Injury Characteristics

3.1 Injury Onset

Given that football is a collision sport, it would be expected that most football injuries are acute, as opposed to overuse or gradual onset injuries. None of the studies reviewed here directly addressed this issue, so no conclusions or comparisons can be drawn. The only hint along these lines comes from some unpublished data from Zemper on college football, which indicates that over a 4-year period only 5% of 4,559 reported injuries (that kept a player out for one day or more) were overuse or gradual onset injuries.

3.2 Injury Type

A review of the literature summarized in Table 4.3 indicates that at both the high school and college levels the three most commonly occurring types of injuries in the game of football are sprains (ligaments), strains (muscles), and contusions. This result was fairly consistent across all the studies, including the only two college level studies that provided a breakdown of injuries by type. The great discrepancies seen in the percentages of sprains and strains reported by Lackland (52) raise the question of how these two injury types were defined in this particular study, but also may reflect other differences between this and other studies, such as definitions of a reportable injury and methods of reporting and collecting data.
______________________________________________________________________________________
Table 4.3 A Percent Comparison of Injury Types
______________________________________________________________________________________

______________________________________________________________________________________
Information about injury type and location is important because it allows all parties involved in the game of football to pay particular attention to these areas so that effective training, conditioning, preventive, and rehabilitative measures can be developed and employed.

3.3 Injury Location

3.3.1 Head/Spine/Trunk

The head (skull and brain) is particularly susceptible to injury because it cannot be conditioned to accept trauma, and once injured is further susceptible to injury (5, 22, 35, 50). Cantu (22) describes three principles to understand how biomechanical forces produce head injury: (a) Coup injuries occur when the head in a resting state is struck by another object producing maximal brain injury beneath the point of cranial impact; (b) contra-coup injuries occur when the head collides with a nonmoving object producing maximal brain injury opposite the side of cranial impact; and (c) direct injuries to the brain tissue may occur in the presence of a skull fracture when the bone displaces at the moment of impact.
Cantu (22) also lists several other types of head injuries that may occur as a result of football. These include intracranial hemorrhage: (a) epidural hematoma, (b) subdural hematoma, (c) intracerebral hematoma, and (d) subarachnoid hematoma; and malignant brain edema syndrome.
In both coup and contra-coup injuries, shearing forces (a force applied parallel to a surface) appear to be the major type of stress generated by an applied force. Two less severe types of stress are also generated by an applied force. These include compressive and tensile (negative pressure) stresses (22).
Head injuries also may occur as the result of an indirect (glancing) blow to the head resulting in impulsive loading in which the head is set into motion (22). Concussions may occur when "the skull is put into motion before the contained brain" (74).
Many definitions of a concussion exist in the literature, but there does seem to be some consensus that concussions are either transient or protracted and are the result of traumatically induced alterations, impairments, and dysfunction of mental (neural) function with manifested signs and symptoms including one or more of the following: loss of awareness, amnesia, dizziness, headaches, blurred vision, double vision, loss of equilibrium, feelings of deja vu, and visual and auditory hallucinations (14, 22, 35, 50).
A review of the literature indicates that cerebral concussions are the most frequently occurring types of head injuries in the game of football at the high school and college levels (3, 13, 22, 54, 104, 106). As shown in Table 4.3, concussions account for at least 5% of all reported football injuries, or at least one out of every 20 injuries.
The high incidence of cerebral concussion in the game of football is compounded by the fact that athletes appear to be susceptible to more severe head injury following an initial concussive event in which symptoms from the prior injury persist. In the literature this has been labeled as the second impact syndrome (22, 81). Furthermore, studies have shown that players suffering an initial concussion are four times more likely to sustain a second concussion compared with a player with no prior concussion (22, 35). The study of high school players by Gerberich et al. (35) involved only loss of consciousness injuries in arriving at their finding of four times greater risk for a subsequent injury. A recent study of college football players by Zemper (109) showed that players with a history of cerebral concussion (of any severity) any time during the previous 5 years were six times as likely to suffer a new concussion as those with no history. Albright et al. (3) documented that 26 of 78 players with an initial head injury had a subsequent head injury during their collegiate career, and these were more severe than the initial injury. In the same vein, Cantu (22) stressed that athletes are less likely to suffer subsequent concussions if they have fully recovered from the initial cerebral concussion.
There appears to be some controversy in the literature concerning the mechanisms involved in cervical neck/spine injuries occurring in football. Hyperflexion, hyperextension, and axial loading have all been proposed as the primary mechanism contributing to cervical injuries.
Injuries to the cervical spine traditionally have been attributed to hyperflexion and/or hyperextension mechanisms (23, 32, 33, 79, 82). Hyperflexion, in which the head is driven downward (as seen in head butting), may be the more dangerous and frequent mechanism involved in cervical spine injuries (33, 55). In a retrospective analysis conducted by Maroon (55) seven neck injuries involved hyperflexion as the primary mechanism of injury. Five of the seven players injured were involved in the process of tackling.
Schneider et al. (82) suggested that cervical spine injuries documented in their study resulted from hyperextension (shear forces) when the face mask was impacted, driving the posterior rim of the helmet against the back of the neck and providing a "guillotine" effect. This suggestion led Schneider et al. (82) to recommend that the rim of the helmet be cut higher to prevent its impact on the neck. However, this suggestion has been refuted, since more dangerous hyperextension occurs to the upper cervical spine when the helmet is cut high. A reduction in force is seen when the posterior rim of the helmet is allowed to make contact with the back of the neck and even further loading conditions are evident when the posterior rim of the helmet makes contact with the shoulder pads (23).
Axial loading occurs when the neck is slightly flexed, cervical lordosis is straightened, and the spine is converted into a segmented column. When a load is applied to the top of the helmet, as in spearing, force is transmitted down the straightened cervical spine, and an axial load is created and transmitted to the spinal structures. This results in the spine being compressed and crushed, and fractures and/or dislocations will occur (96).
Injuries to the cervical spine can be divided into four categories: (a) sprains, (b) strains, (c) fractures, and (d) dislocations (7). Torg (95) further categorized spinal injuries into five groups according to their severity. Group I is made up of mild sprain and strain injuries that improve with or without treatment, and includes burners or stingers (pinched cervical nerve syndrome). Group II is made up of moderate injuries and includes compression fractures, degenerative spine changes, and brachial plexus axonotmesis. Group III is made up of more severe neck injuries and includes unstable lesions with fractures, dislocations, and/or subluxation without neurologic deficit. Group IV is made up of very severe injuries including fracture-dislocations with neurologic deficit. Group V is made up of catastrophic disabling injuries, which include permanent quadriplegia or death. We will discuss the latter group in section 4.2--Catastrophic Injuries.
Lateral flexion (stretching, bending, or traction of the cervical nerve roots) has been identified as a mechanism of injury in cervical neck/spine injuries. This mechanism results in brachial plexus injuries, commonly called burners or stingers. These occur when a player receives a forceful blow to the head from the side with simultaneous depression of the opposite shoulder. Injury also may result from head extension or by depression of the shoulder while the neck and head are fixed.
A review of the literature indicates that brachial plexus injuries associated with a prickly burning sensation in the shoulders, and pain and weakness in the neck, arms, and hands, are the most commonly observed neck injuries (33, 54, 87). Robertson et al. (77) in a study of collegiate football players indicated that approximately 50% of the players on the team experienced at least one and sometimes more injuries of the plexus during a season. Furthermore, players with repeated incidence of this type of injury were at higher risk of developing further weakness and neurological dysfunction. Dolan et al. (32) report that a collegiate football team may experience an average of six stingers during a single game with as many as five or six major neck injuries, including fractures, disk or ligament disruptions, or persistent neuropathies, during one football season. Albright et al. (3) indicated that the majority (55%) of the neck injuries suffered by collegiate players were of a soft tissue or skeletal nature with sprains being the most common type of injury (64% with a mean time loss of 5.5 days). When reviewing high school incidence rates, Albright et al. (2) found that 32% of college freshman players demonstrated evidence of prior neck injury sustained during their high school football careers.
3.3.2 Upper Extremities

Injuries to the upper extremities commonly occur in the game of football. The most frequently reported cases appear to be the shoulders, hands, and fingers. The frequent occurrence of these injuries is not surprising considering that the hands and fingers are minimally protected by padding and are under constant trauma through blocking and tackling, and although the shoulder is protected by pads, it is often the initial contact point in tackling and blocking. The hand may be particularly vulnerable to injury because it is used often in direct contact against the headgear, the face mask, the shoulder pads, and the opponent's body. Injuries to the fingers also may occur as players grasp the opponent's belts, jerseys, and pads while tackling and blocking.
3.3.3 Lower Extremities

Football, a violent collision and contact sport, puts some of the greatest demands on the lower extremity because it is the major weight-bearing area of the body and is usually fixed to the ground. Additionally, football players are required to make severe running, cutting, and jumping moves, all adding to the stress placed on the lower body. The knee is particularly vulnerable to injury because it is minimally protected by padding and is often in an exposed position (20, 73, 90).
Because knee motion is limited to internal and external rotation as the leg straightens (extension) and bends (flexion), injury may occur when the joint is stressed into awkward and unusual positions, usually as the result of direct contact from other players. Noncontact injuries often involve unnatural twisting motions and drills that prevent the natural rotation of the knee joint. Such injuries are commonly seen in cutting maneuvers when the foot is fixed to the ground, causing damage to the anterior cruciate ligaments, the collateral ligaments, menisci, and patella (30, 90, 99).
A review of Tables 4.4 and 4.5 demonstrates that, at the high school and college levels, the knee is the most frequently injured body part in the lower extremity. Among the most common knee injuries in football are ligament injuries (grade I, grade II, and grade III sprains), of which the medial collateral ligament is the most commonly sprained (31). The frequent occurrence of knee sprains is a concern because a significant percentage of these injuries may require surgical intervention and result in significant time loss from active participation (31, 47). Additionally, injury to the medial collateral ligament often increases the likelihood of further injury to the supporting structures of the knee, particularly damage to the anterior cruciate ligament (70). Other common knee injuries include medial femoral condyle fractures, epiphyseal fractures, anterior and posterior cruciate damage, meniscus injuries, and patellofemoral arthritis (8, 45, 58, 63, 100).
A review of Tables 4.4 and 4.5 shows that injuries to the lower extremities also commonly occur in the ankle, toe, hip, quadriceps, and hamstring. According to Rishel et al. (76), most ankle injuries are inversion sprains that result in tears to the lateral ligament structures. Damage to these ligaments leads to ankle instability, making the athlete more susceptible to recurrent ankle injuries.
Injuries to the toe, particularly the metatarsal-phalangeal joint (bunion joint of the big toe), also are significant lower extremity injuries commonly seen in football players. This injury is referred to as turf-toe (100).
A 5-year retrospective study conducted by Heiser et al. (40) demonstrated that injuries to the hamstring may also be common among collegiate football players, highlighting the need for an evaluation and rehabilitation program to prevent and lessen the frequency and severity of hamstring strains. The study revealed that, prior to a strengthening and rehabilitation program, 7.7% of the players reported hamstring strains, with 31.7% sustaining one or more recurrent injuries. Players utilizing a Cybex II program reported only a 1.1% hamstring injury rate with no recurrent injuries being reported.
______________________________________________________________________________________
Table 4.4 A Percent Comparison of Injury Location
______________________________________________________________________________________

______________________________________________________________________________________

______________________________________________________________________________________
Table 4.5 A Rate Comparison of Injury Location (Injury Rate/1,000 Athlete-Exposures)
______________________________________________________________________________________

High School

___________

___College____

________

Alles et al., 1979

Alles et al., 1979

Buckley & Powell, 1979

Zemper, 1989b

# injuries:

1696

12,432

7190

3744

Study type:

P^a

P

P

P

Injury location____

# participants:

____2674____

___13,416___

___28,419___

____8325___

Head

0.62

0.78

1.7

0.44

Spine/trunk

0.63

0.49

0.9

0.87

Neck/spine

--

--

--

0.32

Trunk (torso)

0.63

0.49

0.9

--

Back

--

--

--

0.27

Chest/rib

--

--

--

0.15

Abdomen

--

--

--

0.02

Groin

--

--

--

0.11

Upper extremities

1.59

1.75

2.5

1.32

Shoulder

0.67

1.07

1.4

0.73

Arm (upper&lower)

0.92

0.68

1.1

0.05

Elbow

--

--

--

0.12

Wrist

--

--

--

0.05

Hand/finger

--

--

--

0.37

Lower extremities

4.61

6.20

8.1

3.34

Pelvis/hip

--

--

--

0.18

Knee

1.95

3.16

3.1

1.19

Leg (upper&lower)

1.19

1.30

2.4

0.85

Ankle

1.47

1.74

2.6

0.93

Foot/toe

--

--

--

0.19

Other

0.40

0.00

0.2

0.08

_______________________________________________________________________________________
^{aP = prospective cohort study.
___________________________________________________________________________________________________

4. Injury Severity
Injury severity for our purposes is broken down into two broad categories. The first is the common time-loss injury from which the athlete recovers in a matter of hours, days, or occasionally weeks, and which usually is measured by number of days before return to unrestricted activity. The second category is the catastrophic injury resulting in some form of permanent disability. This latter category also includes fatalities.

4.1 Time Loss

Few of the articles reviewed for this chapter had more than a minimal mention of time loss in relation to injuries. Blyth and Mueller (10) found in 4 years of covering football injuries in 43 North Carolina high schools that 18.1% of the reported injuries kept a player out for one day, 47.2% for 2 to 6 days, and 34.7% for 7 or more days. At the collegiate level, Buckley (15) found in 5 years of NAIRS data from the late 1970s that the total reportable injury rate (time loss of 1 day or more) was 10.1/1,000 athlete-exposures, whereas injuries severe enough to cause more than 7 days of time loss occurred at a rate of 2.7/1,000 A-E. More recently, data on college football from Zemper (105) (3 years) and the NCAA (67) (6 years) found injury rates of one day or more duration to be 6.1 and 6.4/1,000 A-E, respectively.
Data on college football injuries during the 1988 to 1990 seasons collected by Zemper (unpublished) included the number of days lost for each of 3,363 injuries reported. For injuries that kept a player out for 1 to 2 days the rate was 1.7/1,000 A-E (27.6% of the total number of injuries), for 3 to 4 days 1.3/1,000 A-E (20.9%), 5 to 6 days 0.8/1,000 A-E (12.2%), 7 to 9 days 0.7/1,000 A-E (10.4%), and for 10 or more days 1.8/1,000 A-E (28.9%). For the 3,363 injuries there were a total of 48,730 days lost, with an average of 14.6 days lost per injury.

4.2 Catastrophic Injury
Catastrophic injuries are defined as injuries that result in death or some type of permanent disability. The nonfatal injuries generally include permanent paralysis (quadriplegia, paraplegia, or hemiplegia) caused by spinal cord damage or brain trauma (e.g., subdural hematoma).
4.2.1 Spinal Cord

Reference to Table 4.6 indicates that during the 16-year period from 1977 to 1992 a total of 155 football players had incomplete neurological recovery from cervical cord injuries (62). One hundred and twenty-seven of these injuries were to high school players, and 20 to college players. The remaining 8 injuries were to players participating at other levels (sandlot and professional). Although the 1988, 1989, and 1990 data suggest an increase in cervical cord injuries, the 1991 data show the most dramatic reduction since the beginning of the study in 1977. In 1991 there was only one cervical cord injury with incomplete neurological recovery at the high school level and none at the college level.
Table 4.7 illustrates the incidence rates of spinal cord injuries for both high school and college participants. The incidence rates per 100,000 participants are low in both high school and college. Based on the 16 years covered in this table, the high school incidence is 0.59 per 100,000 participants and the college incidence is 1.66 per 100,000 participants.
When comparing cervical cord injuries in offensive and defensive players, it appears to be safer playing offensive football. From 1977 to 1992, 113 of the 155 players (73%) with cervical cord injuries were playing defense with only 24 of the injuries occurring on the offensive side of the ball. A majority of the defensive players were injured as a result of tackling, which clearly is the most dangerous activity in the game of football. Past yearly reports by Mueller et al. have revealed that defensive backs were injured at a higher rate than other positions.
_______________________________________________________________________________________
Table 4.6 Football Fatalities and Catastrophic Injuries (1970-1992)
_______________________________________________________________________________________

_________

High School

_________

________

___College__

_______

Year_______

Direct __fatalities__

Indirect __fatalities__

Incomplete _recovery_

Direct __fatalities__

Indirect __fatalities__

Incomplete _recovery_

1970

23

12

--

3

2

--

1971

15

7

--

3

2

--

1972

16

10

--

2

1

--

1973

7

5

--

0

3

--

1974

10

5

--

1

3

--

1975

13

3

--

1

3

--

1976

15

7

--

--

2

--

1977

8

6

10

1

0

2

1978

9

8

13

0

1

0

1979

3

8

8

1

1

3

1980

9

4

11

0

0

2

1981

5

6

6

2

0

2

1982

7

7

7

0

3

2

1983

4

6

11

0

3

1

1984

4

3

5

1

0

0

1985

4

1

6

1

1

3

1986

11

7

3

1

1

0

1987

4

4

9

0

3

0

1988

7

10

10

0

0

1

1989

4

9

12

0

2

2

1990

0

3

11

0

3

2

1991

3

3

1

0

1

0

1992

2

8

4

0

1

0

Totals

183

142

127

17

36

20

_______________________________________________________________________________________
Note: Figures are updated annually to include new cases investigated after publication. Catastrophic injuries are defined as those involving some disability (monoplegia, hemiplegia, paraplegia, quadriplegia) at the time of injury, including complete or incomplete neurological recovery. In this table incomplete recovery was the result of cervical cord injuries. Direct fatalities and injuries are defined as those resulting directly from participating in the game of football. Indirect fatalities are defined as those resulting from systemic failure as a result of exertion while participating in the game of football or by a complication secondary to a nonfatal injury.
____________________________________________________________________________________________________

_______________________________________________________________________________________
Table 4.7 Direct Football Fatalities and Catastrophic Injury Incidence per 100,000 Participants (1970-92)
_______________________________________________________________________________________

______High School_____

_____College_____

Year____

Fatality incidence

Catastrophic incidence

Fatality incidence

Catastrophic incidence

1970

1.92

--

4.00

0.00

1971

1.25

--

4.00

0.00

1972

1.33

--

2.67

--

1973

0.58

--

0.00

--

1974

8.83

--

1.33

--

1975

1.08

--

1.33

--

1976

1.00

--

0.00

--

1977

0.53

0.77

1.33

2.67

1978

0.60

1.00

0.00

0.00

1979

0.23

0.62

1.33

4.00

1980

0.69

0.85

0.00

2.67

1981

0.38

0.46

2.67

2.67

1982

0.54

0.54

0.00

2.67

1983

0.30

0.85

0.00

1.33

1984

0.30

0.38

1.33

0.00

1985

0.30

0.46

1.33

1.00

1986

0.77

0.23

1.33

0.00

1987

0.30

0.69

0.00

0.00

1988

0.46

0.77

0.00

1.33

1989

0.27

0.80

0.00

2.66

1990

0.00

0.73

0.00

2.66

1991

0.20

0.07

0.00

0.00

1992

0.07

0.27

0.00

0.00

_______________________________________________________________________________________
Note: Fatality incidence and catastrophic incidence are based on 1,500,000 junior and senior high school players and 75,000 college players.
_______________________________________________________________________________________

4.2.2 Fatalities

Table 4.6 shows that after a slight rise in the number of football fatalities during the 1986 season, the 1990 data revealed the elimination of direct football fatalities (62). That was the first time in 59 years that there have been no direct football fatalities.
Since 1960 most of the direct fatalities have been caused by head and neck injuries. The authors of the Annual Survey of Football Injury Research, Dr. Mueller and Mr. Schindler, are convinced that current rules eliminating the head in blocking and tackling, the helmet research conducted by the National Operating Committee on Standards for Athletic Equipment (NOCSAE), excellent physical conditioning, proper medical supervision, and a good data collection system have played primary roles in reducing fatalities and serious head and neck injuries in football. This is illustrated by the increase in both head and cervical spine fatalities during the decade from 1965 to 1974. This time period was associated with blocking and tackling techniques that involved the head as the initial point of contact. In the decade from 1975 to 1984 there was a reduction in head and cervical spine injuries that was associated with the 1976 rule change eliminating the head as the initial contact point in blocking and tackling. There is no doubt that the 1976 rule change has made a difference, and that a continued effort should be made to keep the head out of the fundamental skills of football. These data illustrate the importance of data collection and analysis in making changes in the game of football that help reduce the incidence of serious injuries.
A majority of the indirect fatalities are heat related (62). A continuous effort should be made to eliminate heat stroke deaths associated with football. There have been 46 heat stroke deaths from 1970 to 1992, and since 1974 there has been a dramatic reduction in these deaths with the exception of 1978 when there were four. There was one heat stroke death in 1992. All coaches, trainers, and physicians should continue their efforts toward eliminating athletic fatalities that result from inappropriate physical activities in hot weather.

4.3 Clinical Outcome/Residual Symptoms

Very little work on long-term outcome of football injuries currently can be found in the literature; it is an area that needs attention in the future. The only example of this type of research found in this survey was a long-term follow-up study of 23 high school athletes that followed the relationship between knee injuries and the development of knee osteoarthritis. Moretz (60) showed that 39% of the injured football players suffered knee injuries and that half of them showed significant degenerative changes and/or functional disability of their knees.

5. Injury Risk Factors
5.1 Intrinsic Factors

5.1.1 Physical Characteristics

There is evidence in the literature that lower extremity injuries may be the result of lower body strength imbalances and other leg deficiencies. According to Darden (30), 80% of all knee injuries occur to the weaker of the two legs, with as many as 88% of injuries occurring to athletes with leg length inequalities. Furthermore, athletes with insufficient ligament stability as a result of poor muscular support are prone to hyperextensive knee injuries.
Other researchers have provided evidence to dispel the theories of muscle imbalances and ligament instabilities as predisposing factors for lower extremity injury. In a prospective blind study conducted by Grace et al. (36) on high school football players, no relationship was found between isokinetically measured thigh-muscle imbalances and the increased likelihood of knee joint injuries. In a similar vein, Kalenak et al. (48) reported that selective strengthening and stretching activities, to correct for joint instability in loose-jointed college football players, did not reduce the incidence of knee injuries. Furthermore, tight-jointed and loose-jointed football players had similar knee ligament rupture incidence rates. In one other study, it was shown that ligament laxity (flexibility) tests were ineffective in predicting knee injuries in collegiate football players (59).
There does seem to be some agreement in the literature about the importance of overall lower extremity strength, conditioning, and maintenance. Athletes with poor muscle strength, particularly in the quadriceps, hamstring, and gastrocnemius muscles, are more prone to injury caused by the lack of support these muscles provide to the surrounding structures of the knee and other joints (30, 91, 99). It also has been postulated that athletes with tight heel cords may be more susceptible to lower extremity injury (98).
A review of the literature demonstrates that much research is needed in regard to the relationship size, age, weight, and height play in the overall picture of lower extremity injury. In one study it was shown that players with a greater mean height and weight showed a higher incidence of knee-joint injuries (36). However, from the way the data was analyzed, it was not evident whether the explanation for this could have been that taller, heavier players, who tend to be older, could have been receiving more exposure to the possibility of being injured because they were first string players.
Neck injuries in football can lead to serious problems, and it is clear from the literature that players with neck injuries are at a higher risk for future injuries to this area. In a study conducted 20 years ago, prior to several rule changes regarding blocking and tackling aimed at reducing head and neck injuries, Albright et al. (2) found that one third of college freshman football players demonstrated evidence of prior neck injury sustained during their high school football careers. Similar trends were evidenced for back injuries and head injuries. Albright found that after an initial injury, the chances of a player suffering subsequent head or neck trauma increased to 42% with a 67% chance that the injury would occur in some subsequent season. It also appears that players with long, thin, and poorly developed necks are at a higher risk of sustaining a neck injury (87).
Brachial plexus injuries are the most commonly observed neck injuries, and Robertson et al. (77) found that approximately 50% of the players on a college team experienced at least one and sometimes more injuries of the brachial plexus. Players with this type of neck injury are felt to be at a high risk of developing further weakness and neurological dysfunction.

5.2 Extrinsic Factors

5.2.1 Exposure
The inherently violent nature of the game of football (full contact drills and activities) combined with all the physically demanding aspects of the game (running, jumping, diving, tackling, blocking, and cutting maneuvers) make it easy to understand why and how football players are so prone to injury.
A review of the literature reveals several factors that contribute to the incidence of injury among players at different positions. (See Tables 4.8 and 4.9 for a comparison of injuries by player positions.) First, there appears to be a direct relationship between the amount of contact a player gives or receives and the incidence of injury. Whiteside et al. (102) state that the offensive and defensive line players are at greatest risk for injury because they are involved in contact on every play. In their study, offensive and defensive line players sustained 47% of the total number of time-loss injuries and illnesses with the offensive line exhibiting the largest percentage (29%).
_______________________________________________________________________________________
Table 4.8 A Comparison of Injury by Player Position in High School Football
_______________________________________________________________________________________

Culpepper, & Nieman, 1983

Blyth & Mueller, 1974

Olson, 1979

Prager, 1989

NATA, 1987^a

Type of study:

Case series

Prospective

Prospective

Prospective

Prospective

Player

Type of rate:

% of injuries

% of injuries

% of injuries

% of injuries

% of injuries

Inj./100 games

_position_

# injuries:__

1877_____

4287_____

465______

251______

__range__

4292_____

Offense

End

--

--

12.8

--

-- - 12.8

3.4^b

Tackle

11.6

5.7

8.2

19.5^c

5.7-9.5

--

Guard

7.8

7.1

1.9

9.6^c

1.9-9.6

--

Center

4.3

3.6

2.3

5.2^c

2.3-5.2

--

Quarterback

8.7

6.1

3.9

6.4

3.9-8.7

6.9

Running back

19.6

21.1

26.1

15.9

15.9-26.1

9.3

Fl./wide recvr

7.2

3.3

--

2.0

2.0-7.2

2.9

Tight end

5.2

4.7

--

6.4

4.7-6.4

4.2

Totals

64.4

51.6

55.2

65.0

51.6-65.0

26.7

Defense

End

4.7

9.2

8.2

--

4.7-9.2

--

Lineman

5.5

16.7

17.9

--

5.5-17.9

4.7

Def. back

9.2

10.6

16.3^d

7.2

7.2-16.3

3.4

Linebacker

9.0

11.4

--

13.1

9.0-13.1

5.1

Half/cornerback

--

--

--

--

--

--

Safety

--

--

1.6

2.0

1.6-2.0

--

Totals

28.4

47.9

44.0

22.3

22.3-47.9

13.2

Kicker/punter

--

--

0.8

1.6

0.8-1.6

--

Other

7.2

0.2

--

11.1

0.2-11.1

--

Totals

7.2

0.2

0.8

12.7

0.2-12.7

--

_______________________________________________________________________________________
^{aGame-related injuries only.
^{bListed as injuries to offensive line.
^{cNo differentiation made between offensive and defensive linemen.
^{dListed as injuries to defensive backfield.
___________________________________________________________________________________________________}}}
_______________________________________________________________________________________
Table 4.9 A Comparison of Injury by Player Position in College Football
_______________________________________________________________________________________

Canale, 1981

Whiteside, 1985

NCAA ISS, 1984

NCAA ISS, 1985

Zemper, 1989b

Type of study:

Prospective

Prospective

Prospective

Prospective

Prospective

Player

Type of rate:

% of injuries

% of injuries

Per 1,000 A-E

Per 1,000 A-E

Per 1,000 A-E

_position_

# injuries:__

283______

2186_____

3218_____

3002_____

3744_____

Offense

End

7.8^a

28.9^a

0.81

0.81

0.30^b

Tackle

--

--

1.03

0.52

0.22

Guard

--

--

1.13

0.57

0.23

Center

--

--

0.46

0.46

0.18

Quarterback

5.9

1.3

0.58

0.58

0.26

Running back

21.6

14.3^c

1.88

0.95

0.43

Slot/wingback

--

--

0.16

0.16

0.06

Fl./wide recvr

7.8

6.3

1.03

1.03

0.48

Tight end

3.9

--

--

--

--

Totals

47.0%

50.8%

7.08

5.08

3.02

Defense

End

11.8

--

--

--

--

Lineman

17.6

17.9

2.17

0.43

0.23

Def. back

21.6

14.0^d

^--

--

--

Linebacker

2.0

14.6

2.04

0.68

0.32

Half/cornerback

--

--

1.36

0.69

0.29

Safety

--

--

0.82

0.82

0.15

Totals

53.0%

46.5%

6.39

2.62

0.99

Special teams

--

--

0.39

0.39

--

Kicker/punter

--

2.7

0.11

0.11

--

Other

--

--

0.05

0.05

--

Totals

--

2.7%

0.55

0.55

--

_______________________________________________________________________________________
^{aListed as injuries to offensive line.
^{bAdjusted for number of players at each position (e.g., rate for offensive tackles is 0.44 per 1,000 A-E and for centers is 0.18, but because there are two tackles and one center playing at a time, the adjusted rates are 0.22 and 0.18, respectively.
^{cListed as injuries to offensive backs.
^{dListed as injuries to defensive backs.
___________________________________________________________________________________________________}}}Powell (73) states that the high incidence of injury to offensive and defensive linemen may be due to the fact there are more linemen on the field at any one time than other players. When adjusting for each individual player at his position, running backs evidenced the greater incidence of injury with 9.3 injuries per 100 games. Offensive and defensive linemen combined incurred 8.0 injuries per 100 games.
Further review of the literature indicates that the style of play exhibited by any particular team may also be a factor as to which players are going to be most prone to injury. As for practice situations, a study by Cahill et al. (17) revealed that contact activities were 4.7 times more likely to produce an injury than controlled activities. Mueller and Blyth (61) also found that limited contact practices yielded significantly lower rates of injury compared with practices with regular contact.
An interesting finding in the literature is that football players in high school will play more than one position on both offense and defense (72). It then would seem that high school athletes are at a greater risk for injury than collegiate players because of these higher exposure rates and resulting fatigue factors. However, no conclusions were drawn from this finding.
In a prospective study on a group of collegiate football players, Derscheid et al. (31) showed that there was a relationship (although small) between player position and the risk of reinjuring the knee. Defensive team members, particularly linebackers, middle guards, and defensive tackles, appeared to be at higher risk for knee injury.
Some studies have shown that the largest proportion of knee injuries occurs in practice and scrimmage situations. Cahill et al. (18) states that nearly 55% of knee injuries occur in nongame-related activities compared with nearly 44% during games. These results are supported by Derscheid et al. (31), who found knee injuries were more than three times as likely to occur in spring practices, with their greater frequency of scrimmages, compared with the fall, when rates were based on player exposures. Keeping in mind the difference between the percentage or proportion of total injuries and an injury rate based on amount of exposure, one can postulate that the greater number of injuries during practices and scrimmages is linked to the greater number of hours involved in training and practice (exposure time), as well as the larger numbers of players involved in these situations.
5.2.2 Training Methods

The lack of a well-rounded full year conditioning and rehabilitative program also may be viewed as a potential risk factor. A conditioning program with emphasis on stretching and strength training should be implemented at both the high school and college levels. Rehabilitative programs need to focus on reconditioning athletes to full strength before they return to practice and competition. Additionally, the lack of adequate preparticipation examinations, improper injury recognition, screening, and evaluation may all be considered potential risk factors (99).
An 8-year study of high school football varsity players demonstrated that preseason conditioning of the total body decreased the incidence of early season knee injuries, decreased the total number of knee injuries throughout the season, and decreased the severity of injuries. Linemen experienced greater benefits from the preseason conditioning program, with a 61% reduction in knee injuries, with improvements also being seen in flexibility and agility (16). In a follow-up study, Cahill et al. (18) demonstrated that even with the reduction of direct supervision in a preseason total body conditioning program, the incidence and severity of knee injuries were still decreased.
The incidence of football injuries may be a direct reflection of the coaching staff and their coaching philosophies and practices. A football program directed by incompetent and inexperienced personnel is more likely to suffer and promote a large number of football injuries. It is, therefore, primarily the coaches' responsibility to properly educate, prepare, condition, and train their athletes with regard to proper playing techniques. Additionally, a program using faulty and inadequate equipment and playing surfaces provides further factors contributing to the increased incidence of football-related injuries (44).
Differences in physical maturity between the players on the field may cause an unfair and unsafe playing environment. Much of the increased frequency of football-related injuries is the result of pitting players of varying size, strength, speed, and agility against each other (101).
5.2.3 Environment

Fixation of the foot through rigid cleating has been shown to be a primary factor in the production of lower extremity injuries, particularly of the knee and ankle (10, 20,42). In a study conducted by Torg et al. (94) it was demonstrated that the conventional football shoe with seven 3/4 inch cleats caused the foot to become excessively fixed to the ground and uncompliant to any movement forces. In contrast, a soccer-style shoe with multiple, shorter cleats reduced foot fixation and resulted in fewer and less severe knee injuries. On the basis of these results, Torg et al. (94) recommended replacing the conventional football shoe with shoes of the following specifications: (a) synthetic molded soles, (b) minimum of 15 cleats per shoe, (c) minimum cleat diameter of 1/2 inch, and (d) maximum cleat length of 3/8 inch.
As a related factor, it has been postulated that the condition of the playing field contributes to the increased risk of lower extremity injury. In 1974 Mueller and Blyth (61) demonstrated in an experimental study that well-maintained playing fields decreased the risk of knee and ankle injuries with a 42.2% reduction in the number of injuries being observed when soccer-style shoes were used on the resurfaced fields.
The type of playing surface also may be a potential predisposing factor for lower extremity injury. In an effort to decrease foot fixation, artificial turf was developed, but a review of the literature shows a number of studies indicating this type of surface increases the number of time-loss lower extremity injuries by 30% to 50% (4, 20, 41, 88).
It is important to note that several studies have disputed the findings of increased injury risk from artificial turf. Based on football injury data obtained from 41 college teams during the 1975 season, it was found that "artificial turf did not constitute an imminent hazard" to football teams (24). In support of these findings, Henschen et al. (41) revealed that grass and artificial surfaces produced similar injury rates, but the most serious injuries occurred on artificial turf. Further evidence is provided by Culpepper et al. (29), who found that the probability of sustaining a knee or ankle injury at the high school level was the same for all types of fields.
An interesting finding in the literature revealed that the incidence of injury appears higher in the beginning of the season as well as early in the game for high school teams (10, 28, 38, 43). However, at the collegiate level Zemper (104, 105) reported that the highest rate of game injuries in his study occurred in the third quarter.
5.2.4 Equipment

In recent years there have been two pieces of football protective equipment that have been the focus of research. The helmet has received occasional attention in the literature, and more recently the preventive knee brace has been the subject of a number of articles.
Although a number of articles over the past 25 years have reported the incidence of cerebral concussions, only a few have actually looked at the football helmet. During the 1969 high school football season, Robey et al. (78) found essentially no difference in the incidence of second- or third-degree concussions among brands of padded helmets or brands of suspension helmets. However, the players wearing suspension helmets had lower rates of concussion overall. Data reported by Clarke and Powell (25) from a sample of high school and college teams during the 1975 to 1977 seasons indicated no difference in cerebral concussion rates in 13 brands of helmets. Most recently, Zemper (109) reported on a 5-year study of a national sample of college teams, which indicated that, out of 10 models of helmets studied, one had a statistically significant concussion rate higher than expected and one model had a significantly lower rate of concussions. He concluded that the laboratory testing currently done on football helmets is not sufficient to tell the whole story about the protective capability of helmets, and ongoing epidemiological studies to monitor field performance of helmets and other critical pieces of protective equipment are needed.
In an attempt to prevent the large numbers and severe nature of knee injuries in high school and college football, it is evident from the literature that the most significant trend has been toward the use of preventive (prophylactic) knee braces. A comparison of several studies of the efficacy of the knee brace is shown in Table 4.10.
_______________________________________________________________________________________
Table 4.10 A Comparison of Epidemiologic Knee Brace Studies
_______________________________________________________________________________________

____Study________________Results____________

Grace et al., 1988

Matched pairs; Injury based on time loss from practice; Injuries graded from mild to severe

580 H.S. players over a period of two seasons; 247 wore single-hinged braces; 83 wore double-hinged braces; 250 wore no braces

Significantly more knee injuries in the single-hinged braces group than in double-hinged and nonbraced groups; Foot and ankle injuries more common in braced players; More sever injuries to lower extremity in braced group; More injuries to the knee than other lower extremity areas, MCL tears were most significant knee injury

Hansen et al., 1985

Longitudinal; Number and type of knee surgeries were analyzed; Team medical records were used

University of Southern California (1980-1984)

11% injury rate in nonbraced group, 5% injury rate in braced group; <2% of braced players required surgery, 5% on nonbraced players required surgery; Braced players had one-quarter as many meniscectomies; 1 braced player had severe knee injury, 6 nonbraced players had severe knee injuries; 4% of braced offensive linemen were injured, 18% of nonbraced linemen were injured; 7% of braced defensive linemen were injured, 23% of nonbraced linemen were injured

Hewson et al., 1986

Longitudinal; Team rosters, treatment logs, NAIRS, medical records were analyzed; Injury rates based on number of exposures in games and practices

University of Arizona 1977-1980 no brace use 1981-1985 braces used

Number, type, and severity of knee injuries similar in both groups; Braced and nonbraced players had similar chances of being injured; Knee injury prevention not improved by brace use

Rovere et al., 1987

Longitudinal; Injury records at University Sports Medicine Unit were analyzed; Injury rates based on incidence per 100 players

Wake Forest University 1981-1985 fall seasons and spring practices

More injuries, particularly Grade I MCL sprains, during the braced period; More knee operations performed during the braced period; Regardless of brace use, offensive team suffered more injuries; Knee injuries caused by body contact were fewer during braced period; More injuries occurred during games, with higher rates seen during the braced period

Schriner, 1985

Survey; Based on rates per 100 players; Time-loss parameters used; Injuries classified by injury mechanism involved

1,246 players from Michigan high schools; 1984 season; 197 players wore braces

Knee braces reduced injuries to the medial structures of the knee; No MCL, medial meniscus, and ACL injuries from lateral forces seen in braced players; Braced players did incur hyperextension knee injuries

Shaw and Brubaker, 1987

Longitudinal; Criteria for bracing and injury not defined

Texas City High School varsity and junior varsity; 1983-1986 seasons; 1983 no braces used; 1984 all varsity players wore braces; 1985-86 all players fitted with braces, but not required to wear them

In 1983, 19 knee injuries with 143 days of time loss; 1984 - 17 injuries, 5 days time loss for braced players, 90 days for nonbraced; 1985 - 9 injuries, 30 days time loss for braced players, 22 days for nonbraced; No associated structural damage to players wearing braces

Teitz et al., 1987

Survey; criteria for bracing, exposure while braced, and injury definition not stated

71 NCAA Division I teams in 1984; 61 NCAA Division I teams in 1985; 7,010 players braced, 4,584 players not braced

Braced players had higher injury rates, particularly running backs and defensive backs; MCL injuries were most common; Braced players missed as much or more playing time than nonbraced players; No difference in injury severity between braced and nonbraced players

Zemper, 1990

2-year prospective cohort study; Injury rates based on athlete-exposures; Brace use criteria determined by each school; One day or more time loss needed for injury to be reported

1986 season: 32 NCAA and NAIA teams, 3431 players; 1987 season: 27 NCAA and NAIA teams, 2798 players; 28% of players wore braces

Braces do not reduce MCL injury rate; More time loss for those wearing braces; Approximately half of reported knee injuries occurred in games; No significant difference in knee surgery rate for braced and nonbraced players; No difference between the different brands of braces used; No relationship between brace use and ankle injuries; More MCL injuries among braced players who played on artificial turf; No significant effect of braces in reducing the severity of knee injuries

_______________________________________________________________________________________
A review of this table suggests that despite the wide use of these braces on high school and college football teams, there is little agreement among researchers about their effectiveness in preventing knee injury. Much of the skepticism and inconsistent results stems from the different study designs and methodologies implemented by each researcher. Studies by Hansen et al. (39), Hewson et al. (42), Revere et al. (80), and Shaw and Brubaker (86) were longitudinal in design with distinct brace and nonbrace periods, but each had different criteria for brace use and compliance.
In an attempt to control for potentially confounding variables, Grace et al. (37) matched braced and unbraced players of similar position and size. In a similar vein, Teitz et al. (93) controlled factors such as skill level and position by analyzing different subgroups separately. Most of these studies are limited to one team or a small number of teams, which provides a very small sample size, and these studies provide conflicting results. The only national-scale studies using large numbers of teams were those of Teitz et al. (93) and Zemper (107). Both of these studies indicated preventive knee braces did not reduce the number or severity of MCL injuries, which these braces are specifically designed to prevent.
It is clear that until a sound epidemiologic model for future knee brace studies can be developed, the controversy over knee brace use as a preventive tool in minimizing the risk for lower extremity injury will continue.

6. Suggestions for Injury Prevention

As a preventive measure, the early detection, rehabilitation, and reevaluation of lower extremity strength imbalances and deficiencies must be stressed. This involves a rigorous strength program to develop the muscles in both limbs, as well as an effective strength testing and rehabilitative program (30, 51). A retrospective study conducted by Heiser et al. (40) demonstrated that early detection of muscle imbalances through isokinetic testing followed by a muscle imbalance rehabilitative program decreased the incidence of primary and recurrent hamstring strains in intercollegiate football players.
Prevention of facial injuries may be accomplished through the use of properly fitting helmets and padded chin straps, which will eliminate helmet rotation. Further decreases in injury may occur by padding the rim of the helmet. To reduce the injuries incurred by object penetration through the face mask, players should wear full-cage face masks. The stricter enforcement of rules, the proper use of protective equipment, and the elimination of illegal blocking and tackling techniques all will go a long way toward prevention and elimination of facial injuries (103).
A review of the literature indicates that oral injuries can be prevented through the use of mouth guards/protectors (49, 56, 85). There are several mechanisms contributing to the decreased incidence of oral trauma with mouth protector use: (a) mouth protectors separate the teeth from the soft tissues, thereby preventing lacerations and bruising during impact; (b) mouth protectors cushion opposing teeth and structures; and (c) mouth protectors cushion and distribute the impact and prevent superior and posterior displacement of the condyles. This helps in the prevention of concussions, cerebral hemorrhage, neck injuries, and possible death.
There appears to be consensus that mouthguards offer some type and degree of protection, but there does not seem to be a consensus on which is the best type. Further research is needed to identify the factors associated with oral injuries and mouth protector use and to identify any other factors being influenced by their use. Research also needs to be directed at identifying which players are most susceptible to oral injury so that sound preventive and protective measures can be employed. Furthermore, research is needed to clarify and identify the most effective type of mouth protector for use in high school and college football.
With regard to other types of injuries, many studies in the literature identify and describe the causes, trends, patterns, and factors that contribute to the incidence of injury in football. A summary of preventive recommendations and measures, with discussion and support for each, is presented in tabular form in Table 4.11 to assist coaches, athletes, physicians, athletic trainers, sports administrators, and others to make informed decisions regarding the reduction and elimination of injuries from the game of football. Through careful implementation of and adherence to these preventive measures, football will become a safer and more enjoyable game for all.
_______________________________________________________________________________________
Table 4.11 Suggestions for Injury Prevention

_______________________________________________________________________________________

____Preventive recommendation____

_________________Discussion and Support__________________

Preparticipation examination (Abdenour and Weir, 1986; Blackwell and McCullagh, 1990; Boynton, 1988; Brooks et al., 1976; Caine and Broekhoff, 1987; Coddington and Troxell, 1980; Lysens et al., 1989; Riley, 1978; Robey et al., 1971; Taylor and Luckstead, 1987)

A comprehensive medical history, physical and mental examination prior to athletic participation should be obtained and administered to all players. Abdenour and Weir (1986) demonstrated that historical information about the athlete was a significant predictor of physical well-being. The examination will then serve as a screening device to detect any preexisting medical abnormalities (cardiovascular, neurological, musculoskeletal, respiratory, etc.) and/or physical weaknesses. It will provide baseline information about the health and injury status of the player and will also aid in the detection of prior injury, which has been shown to increase the chance of further injury. The psychological status of the athlete should also be examined and monitored because such factors as life stress, risk-taking behavior, and competitive anxiety have been shown to increase the likelihood of injury.

Maturity assessment (Caine and Broekhoff, 1987; Robey et al., 1971; Whieldon, 1978)

It appears that the risk of injury is not only related to age, but also to the size, physical, and sexual development of the athlete. Caine and Broekhoff (1987) recommend that players be matched according to their level of maturation (size, strength, power, skill level, fitness, experience) to prevent mismatching unequally developed players, thereby preventing unsafe and unfair play.

Year-round injury and health evaluation

This would provide the athletic program with an effective tracking system to constantly evaluate and identify any predisposing conditions that may make the players more susceptible to further more serious injury.

Adequate conditioning (physical and mental) and training of athletes (Boynton, 1988; Brooks et al., 1976; Lysens et al., 1989; Riley, 1978)

A year-round (preseason, inseason, and postseason) mandatory football-specific conditioning and training program should be aimed at improving muscular and ligament imbalances and weaknesses; coordination and timing; flexibility, mobility, and agility; and cardiovascular and endurance capacities. Cahill (1977) reported that players on a weight training program had a decreased incidence of knee injuries because of increased strength of the surrounding ligaments, muscles, and tendons. The conditioning and training program should be a well-monitored gradual progression program with care taken not to overstress the athletes too soon. Physical contact should not be introduced until a general level of fitness has been established. Boynton (1988) states that conditioning has an added quality in that improvement in physical conditioning will lead to subsequent improvements in the mental attitude of the players and of the team as a whole (e.g., confidence and controlled aggression). An adequate and effective treatment and rehabilitation program should also be an integral part of the overall conditioning program of the athlete.

Limited contact activity (Blyth and Mueller, 1974; Cahill and Griffith, 1979; Robey et al., 1971)

Live contact between players can be limited by providing field equipment to teach and practice certain activities. However, safe techniques need to be established when using this equipment as several studies have shown a relationship between field equipment and the incidence of injury. When comparing injury incidence between limited and regular contact programs, Blyth and Mueller (1974) reported there were fewer injuries in the limited contact program (13.3%) than the regular contact program (86.5%).

Adequate provision of medical attention and treatment for the athletes (Blyth and Mueller, 1974; Boynton, 1988; Lackland et al., 1982; Truxal et al., 1981)

The early and effective diagnosis, evaluation, treatment, monitoring, and subsequent prevention of injuries may be accomplished by adequate on-field medical coverage (physicians, sports medicine personnel, and athletics trainers) at all games and practices. Blyth and Mueller (1974) reported that 2.4% of the injuries were the result of inadequate medical care and extreme heat, with only half of the coaches providing liquid to their players during activity. Lackland et al. (1982) demonstrated the value of athletic trainer/team physician services in their high school football injury study. Proper medical supervision during football participation will ensure that activities are performed under safe playing conditions (e.g., avoiding extreme heat, humidity, and cold).

Education, training, and certification of coaches (Caine and Broekhoff, 1987; Blyth and Mueller, 1974; Boynton, 1988)

All coaches should have adequate coaching and playing experience and be trained and evaluated in the fundamentals of the game. The use and coaching of illegal, unsafe techniques should not be tolerated. Coaches should be required to take educational and certification sport and exercise-related courses/clinics to improve, maintain, and refresh their knowledge of the game. A capable, well-trained assistant coaching staff should be a part of all football programs. Schools with several assistant coaches were shown to have a lower incidence rate than those schools without assistant coaches (Blyth and Mueller, 1974).

Rules enforcement (Amold and Coker, 1977; Blyth and Mueller, 1974; Boynton, 1988)

Coaches and athletes need to abide by the rules of the game, with referees and officials playing an active role in their enforcement. The 3.8% of injuries caused by illegal acts such as clipping, spearing, hitting while down and out-of-bounds, and personal fouls reported in the study by Blyth and Mueller (1974) may have been eliminated with proper rule awareness and enforcement.

Equipment (Blyth and Mueller, 1974; Boynton, 1988; Garrick et al., 1978; Johnson and Ritter, 1975; Schwank and Ryan, 1975; Slagle, 1983).

The use of proven protective equipment needs to be an integral part of all football programs. Additional padding and/or taping for the ribs, thighs, arms, hands, and ankles may help prevent some injuries. Additionally, the management of properly fitting and well-maintained equipment needs to be stressed. In the study directed by Blyth and Mueller (1974) 1.7% of all injuries were caused by ill-fitting, defective, or broken equipment. The fields and playing surfaces need to be inspected and well-maintained. Garrick et al. (1978) recommends that a buffer zone be placed around the periphery of the playing field to prevent injuries occurring as the result of contact with people or objects out-of-bounds. Coaches and players should be prohibited from encouraging the use of protective equipment as offensive weapons. Further research needs to be directed at improving the protective equipment being used by football players.

_______________________________________________________________________________________

7. Suggestions for Further Research

In an effort to continue the improvements being made in the area of high school and college football injury prevention, we present several recommendations for future epidemiologic injury research in this sport. These recommendations are provided to help researchers develop and improve a detailed and extensive body of knowledge relating to football injury at the high school and college levels, in an effort to make the game a safer sport for all involved.
1. An injury surveillance system combined with an accurate data collection system needs to be implemented and utilized at all institutions offering organized, competitive football. These programs should have the full cooperation of local physicians, team doctors, coaches, players, administrators, and any other parties involved with the football participants. Data also need to be collected for youth football, for which there currently is little information available.
2. A universally accepted definition of injury in conjunction with universally accepted injury rates, types, and severity measures needs to be developed. This would allow for injury data to be consistently reported and compared across studies. The most commonly used injury definition of "any football-related injury requiring medical attention and causing a player to limit or miss participation for one or more practices or games" is recommended, along with the use of an injury rate involving amount of exposure, such as injuries per 1,000 athlete-exposures or injuries per 1,000 minutes of exposure.
3. Attention should be directed at conducting studies with large sample sizes, equal groups, and an adequate length of study time to ensure the data are reliable and valid. Inconsistencies across studies along these parameters make comparisons difficult. Further attention also is needed to ensure that accurate and detailed methodologies and statistical analyses are provided across studies.
4. In addition to information on individual injury incidents, mechanism, and severity of injury, information concerning the exposure time for all participants (denominator data for calculating injury rates) must be reported and documented in as much detail as possible to make analysis and comparison across studies more effective. This may be accomplished through a universally accepted questionnaire and injury profile to be filled out in the event of an injury.
5. All injuries, regardless of their mechanisms, types, and/or severity levels, need to be reported by trained and experienced personnel (team physicians, athletic trainers, emergency room personnel) if a clear and accurate picture of the injury situation in football is to be established. Additionally, all injuries should be reported and recorded as soon as they occur to eliminate missing or inaccurate data.
6. All players on the football team, in addition to their amount of exposure to injury, need to be accounted for in the analysis of injuries. This would eliminate the problem of some players incurring a higher rate of injury because of their greater exposure times. Additionally, the population at risk and an adequate comparison (control) group needs to be identified in all studies.
7. There needs to be a greater awareness of and control for potential confounding variables and factors contributing to injury. These include age, size, weight, maturity level, skill level, intensity of competition, styles of play, playing conditions, equipment type, prior injury, and coaching experience and philosophy.
8. A qualified sports medicine team needs to be an integral part of all football programs. This should eliminate any bias that may occur in determining the occurrence of an injury.

References
1. Abdenour, T.; Weir, N. Medical assessment of the prospective student athlete. Athletic Training. 21:122-123; 1986.
2. Albright, J.; Moses, M.; Harley, G.; Feldick, G.; Dolan, K.; Burmeister, L. Nonfatal cervical spine injuries in interscholastic football. JAMA. 236(11):1243-1245; 1976.
3. Albright, J.; McAuley, E.; Martin, R.; Crowley, F.; Foster, D. Head and neck injuries in college football: An eight-year analysis. Am. J. Sp. Med. 13(3): 147-152;1985.
4. Alles, W.; Powell, J.; Buckley, W.; Hunt, F. The National Athletic Injury/Illness Reporting System 3-year findings of high school and college football injuries. J. Orthop. Sp. Phys. Ther. 1(2):103-108; 1979.
5. Alves, W.; Rimel, R.; Nelson, W. University of Virginia prospective study of football-induced minor head injury: Status report. Clin. Sp. Med. 6(1):211-218; 1987.
6. Amold, J.; Coker, T. New football rules and athletic injuries. J. Ark. Med. Soc. 74(4):163-165; 1977.
7. Bailey, R. Head, neck and shoulder injuries. Texas Coach. 26-29; 1979.
8. Bandy, W.; McLaughlin, T.; McKitrick, B. A case study: The importance of dynamic evaluation in the assessment of posterior knee pain. J. Orthop. Sp. Phys. Ther. 5:132-133; 1983.
9. Blackwell, B.; McCullagh, P. The relationship of athletic injury to life stress, competitive anxiety and coping resources. Athletic Training. 25(1):23-27; 1990.
10. Blyth, C.; Mueller, F. When and where players get hurt. Physician Sp. Med. 2(9):45-52; 1974.
11. Boynton, D. A model for the prevention of football related injuries: A report to the Ontario Amateur Football Association. Audible. 14:20-23; 1988.
12. Brooks, W.; Young, B. High school football injuries: Prevention of injury to the central nervous system. Southern Med. J. 69(10):1258-1260; 1976.
13. Buckley, W. Concussion injury in college football: An eight-year overview. Athletic Training. 21(3):207-211; 1986.
14. Buckley, W. Concussions in college football: A multivariate analysis. Am. J. Sp. Med. 16(1):51-56; 1988.
15. Buckley, W.; Powell, J. NAIRS: An epidemiological overview of the severity of injury in college football 1975-1980 seasons. Athletic Training. 17:279-282; 1982.
16. Cahill, 8.; Griffith, E. Effect of preseason conditioning on the incidence and severity of high school football knee injuries. Am. J. Sp. Med. 6(4):180-184; 1978.
17. Cahill, B.; Griffith, E. Exposure to injury in major college football. Am. J. Sp. Med. 7(3):183-185; 1979.
18. Cahill, B.; Griffith, E.; Sunderlin, J.; Madden, T.; Weltman, A. High school football knee injuries. Ill. Med. J. 166(5):356-358; 1984.
19. Caine, D.; Broekhoff, J. Maturity assessment: A viable preventive measure against physical and psychological insult to the young athlete. Physician Sp. Med. 15(3):67-80; 1987.
20. Cameron, B.; Davis, O. The swivel football shoe: A controlled study. Am. J. Sp. Med. 16-27; 1973.
21. Canale, S.; Cantler, E.; Sisk, D.; Freeman, B. A chronicle of injuries of an American intercollegiate football team. Am. J. Sp. Med. 9(6):384-389; 1981.
22. Cantu, R. Cerebral concussion in sports. Sports Med. 14(1):65-74; 1992.
23. Carter, D.; Frankel, V. Biomechanics of hyperextension injuries to the cervical spine in football. Am. J. Sp Med. 8(5):302-307; 1980.
24. Clarke, K.; Miller, S. Turf related injuries in college football and soccer. Athletic Training. 12(1):28-32; 1977.
25. Clarke, K.; Powell, J. Football helmets and neurotrauma--An epidemiological overview of three seasons. Med. Sci. Sp. 11(2):138-145; 1979.
26. Coddington, R.; Troxell, T. The effect of emotional factors on football injury rates: A pilot study. J. Human Stress. 6(4):3-5; 1980.
27. Culpepper, M.; Niemann, K. High school football injuries in Birmingham, Alabama. Southern Med. J.· 76(7):873-878; 1983.
28. Culpepper, M.; Niemann, K. A comparison of game and practice injuries in high school football. Physician Sp. Med. 11(10):117-122; 1983.
29. Culpepper, M.; Morrison, T. High school football game injuries from four Birmingham municipal fields. Ala. J. Med. Sci. 24(4):378-382; 1987.
30. Darden, E., Prevention of the knee injury. Audible. 4(1):30-32; 1978.
31. Derscheid, G.; Garrick, J. Medial collateral ligament injuries in football. Am. J. Sp. Med. 9(6):365-368; 1981.
32. Dolan, K.; Feldick, H.; Albright, J.; Moses, J. Neck injuries in football players. AFP. 12(6):86-91; 1975.
33. Funk, F.; Wells, R. Injuries of the cervical spine in football. Clin. Orthop. 109:50-58; 1975.
34. Garrick, J.; Requa, R. Injuries in high school sports. Pediatrics. 60(3):465-469; 1978.
35. Gerberich, S.; Priest, J.; Been, J.; Staub, C.; Maxwell, R. Concussion incidences and severity in secondary school varsity football players. Am. J. Publ. Health. 73(12):1370-1375; 1983.
36. Grace, T.; Sweetser, E.; Nelson, M.; Ydens, L.; Skipper, B. Isokinetic muscle imbalance and knee-joint injuries. J. Bone Joint Surg. 66-A(5):734-740; 1984.
37. Grace, T.; Skipper, B.; Newberry, J.; Nelson, M.; Sweetser, E.; Rothman, M. Prophylactic knee braces and injury to the lower extremity. J. Bone Joint Surg. 70-A(3):422-427; 1988.
38. Hale, R.; Mitchell, W. Football injuries in Hawaii 1979. Hawaii Med. J. 40(7):180-182; 1981.
39. Hansen, B.; Ward, J.; Diehl, R. The preventive use of the Anderson Knee Stabler in football. Physician Sp. Med. 13(9):75-77; 1985.
40. Heiser, T.; Weber, J.; Sullivan, G.; Glare, P.; Jacobs, R. Prophylaxis and management of hamstring muscle injuries in intercollegiate football players. Am. J. Sp. Med. 12(5):368-370; 1984.
41. Henschen, K.; Hell, J.; Bean, B.; Crain, S. Football injuries: Is grass or astroturf the culprit? Utah J. HPERD. 21:5-6; 1989.
42. Hewson, G.; Mendini, R.; Wang, J. Prophylactic knee bracing in college football. Am. J. Sp. Med. 14(4):262-266; 1986.
43. Hoffman, M.; Lyman, K. Medical needs at high school football games in Milwaukee. J. Orthop. Sp. Phys. Ther. 10(5):167-171; 1988.
44. Howe, D. Serious neck injuries. Coaching Review. 1979.
45. Johnson, D. Post-season knee injuries. Journal Clinics. 4:8; 1974.
46. Johnson, J.; Ritter, M. A computer analysis of football injuries. Sp. Med. 3(4):168-171; 1975.
47. Jones, R.; Henley, M.; Francis, P. Nonoperative management of isolated grade III collateral ligament injury in high school football players. Clin. Orthop. 213:137-140; 1986.
48. Kalenak, A.; Morehouse, C. Knee stability and knee ligament injuries. JAMA. 234(11):1143-1145; 1975.
49. Kaufman, R.; Kaufman, A. An experimental study on the effects of the MORA on football players. Basal Facts. 6(4):119-126; 1984.
50. Kelly, J.; Nichols, J.; Filley, C.; Lillehei, K.; Rubenstein, D.; Kleinschmidt-DeMasters, B. Concussion in sports. JAMA. 266(20):2867-2869; 1991.
51. Knight, K. Strength imbalance and knee injury. Physician Sp. Med. 8(1):140; 1980.
52. Lackland, D.; Akers, P.; Hirata, I. High school football injuries in South Carolina: A computerized survey. J. So. Carol. Med. Assoc. 78(2):75-78; 1982.
53. Lysens, R.; Ostyn, M.; Auweele, Y.; Lefevre, J.; Vuylsteke, M.; Renson, L. The accident-prone and overuse-prone profiles of the young athlete. Am. J. Sp. Med. 17(5):612-619; 1989.
54. Maroon, J.; Steele, P.; Berlin, R. Football head and neck injuries: An update. J. Clin. Neurosurg. 27:414-429; 1980.
55. Maroon, J. Catastrophic neck injuries from football in western Pennsylvania. Physician Sp. Med. 9(11):83-86; 1981.
56. McNutt, T.; Shannon, S.; Wright, J.; Feinstein, R. Oral trauma in adolescent athletes: A study of mouth protectors. Fed. Dent. 11(3):209-213; 1989.
57. Meeuwisse, W.; Fowler, P. Frequency and predictability of sports injuries in intercollegiate athletics. Can. J. Sp. Med. Sci. 13(1):35-42; 1988.
58. Moran, M.; Dvonch, V. Subtle Salter Type II distal femoral epiphyseal fracture. Orthopedics. 8(11):1414-1416; 1985.
59. Moretz, T.; Waiters, R.; Smith, L. Flexibility as a predictor of knee injuries in college football players. Physician Sp. Med. 10(7):93-97; 1982.
60. Moretz, J.; Harlan, S.; Goodrich, J.; Waiters, J. Long-term followup of knee injuries in high school football players. Am. J. Sp. Med. 12(4):298-300; 1984.
61. Mueller, F.; Blyth, C. North Carolina high school football injury study: Equipment and prevention. J Sp Med. 2(1):1-10; 1974.
62. Mueller, F.; Schindler, R. Annual survey of football injury research 1931-1992. Orlando, FL: American Football Coaches Association; 1993.
63. Murray, P. Case study: Rehabilitation of a collegiate football placekicker with patellofemoral arthritis. J. Orthop. So. Phys. Ther. 10(6):224-227; 1988.
64. National Athletic Trainers Association. National high school injury registry report. Greenville, SC: NATA; 1987.
65. National Collegiate Athletic Association. Injury Surveillance System: Football. Overland Park, KS: NCAA; 1984.
66. National Collegiate Athletic Association. Injury Surveillance System: Football. Overland Park, KS: NCAA; 1985.
67. National Collegiate Athletic Association. Injury Surveillance System: Football. Overland Park, KS: NCAA; 1990.
68. National Collegiate Athletic Association. Annual Sports Sponsorship Survey. Overland Park, KS: NCAA; 1993.
69. National Federation of State High School Associations. Sports Participation Survey. Kansas City, MO: NFSHSA; 1993.
70. O'Donoghue, D. Diagnosis and treatment of injury to the anterior cruciate ligament. J. Orthop. Sp. Phys. Ther. 2(3):100-107; 1981.
71. Olson, O. The Spokane study: High school football injuries. Physician Sp. Med. 7(12):75-82; 1979.
72. Prager, B.; Fitton, W.; Cahill, B.; Olson, G. High school football injuries: A prospective study and pitfalls of data collection. Am. J. Sp. Med. 17(5):681-685; 1989.
73. Powell, J. Pattern of knee injuries associated with college football 1975-1982. Athletic Training. 20(2):104-109; 1985.
74. Reid, S.; Reid, S., Jr. Advances in sports medicine: Prevention of head and neck injuries in football. Surg. Annual. 13:251-270; 1981.
75. Riley, C. The vital role the school plays in preventing athletic injuries. J. School Health. 1978.
76. Rishel, G.; Wilson, M.; Brodell, J. The management of severe recurrent ankle sprains in a starting high school running back. Athletic Training. 24(3):230-243; 1989.
77. Robertson, W.; Eichman, P.; Clancy, W. Upper trunk brachial plexopathy in football players. JAMA. 241(14):1480-1482; 1979.
78. Robey, J.; Blyth, C.; Mueller, F. Athletic injuries: Application of epidemiological methods. JAMA. 217(2):184-189; 1971.
79. Rogers, B. The mechanics of head and neck trauma to football players. Athletic Training. 16(2):132-135; 1981.
80. Revere, G.; Haupt, H.; Yates, C. Prophylactic knee bracing in college football. Am. J. Sp. Med. 15(2):111-116; 1987.
81. Saunders, R.; Harbaugh, R. The second impact in catastrophic contact-sports head trauma. JAMA. 52(4):538-539; 1984.
82. Schneider, R.; Reifel, E.; Crisler, H.; Oosterbahn, B. erious and fatal football injuries involving the head and spinal cord. JAMA. 177(6):106-111; 1961.
83. Schriner, J.; Schriner, D. The effectiveness of knee bracing in preventing knee injury in high school athletes. Presented at the American College of Sports Medicine annual meeting, Nashville, TN, May 26-29, 1985.
84. Schwank, W.; Ryan, A. Stopping spear tackling. Physician Sp. Med. 3(9):72-78; 1975.
85. Seals, R.; Dorrough, B. Custom mouth protectors: A review of their applications. J. Prosth. Dent. 51(2):238-242; 1984.
86. Shaw, R.; Brubaker, D. The McDavid Guard (MKG) in preventing knee injuries. Texas Coach. 28-29; 1987.
87. Sherk, H.; Watters, W. Neck injuries in football players. J. Med. Soc. New Jersey. 78(9):579-583; 1981.
88. Skovron, M.; Levy, I.; Agel, J. Living with artificial grass: A knowledge update. Am. J. Sp. Med. 18(50):510-513; 1990.
89. Slagle, G. Prevention of neck injuries: The bar roll. Athletic Training. 18(2):63-66; 1983.
90. Slocum, D. The mechanics of common football injuries. JAMA. 170(14):1640-1646; 1959.
91. Tamberelli, A. Prevention and care of the knee injury. Athletic Journal. 58(6):103-105; 1978.
92. Taylor, S.; Luckstead, E. Medical concerns for the adolescent athlete. Iowa Medicine. 1987.
93. Teitz, C.; Hermanson, B.; Kronmal, R.; Diehr, P. Evaluation of the use of braces to prevent injury to the knee in collegiate football players. J. Bone Joint Surg. 69(9):1467-1470; 1987.
94. Torg, J.; Quedenfeld, T.; Landau, S. Football shoes and playing surfaces: From safe to unsafe. Physician Sp. Med. 1(3):51-54; 1973.
95. Torg, T.; Quedenfeld, T.; Moyer, R.; Truer, R.; Spealman, A.; Nichols, C. Severe and catastrophic neck injuries resulting from tackle football. Delaware Med. J. 49(50):267-275; 1977.
96. Torg, J.; Vegso, J.; O'Neill, J.; Sennet, B. The epidemiologic, pathologic, biomechanical, and cinematographic analysis of football-induced cervical spine trauma. Am. J. Sp. Med. 18(1):50-57; 1990.
97. Truxal, B.; Senker, I.; Nussbawn, M. On-field physician assessment of high school football injuries. Physician Sp. Med. 9(9):68-71; 1981.
98. Walsh, W.; Blackburn, T. Prevention of ankle sprains. Am. J. Sp. Med. 5(6):243-245; 1977.
99. Warren, R. Football knee injuries. Coaching Clinic. 20(4):16-21; 1982.
100. Weber, D. Prevention of toe, ankle and knee injuries in football. Athletic Purchasing and Facilities. 6(3):40-41; 1982.
101. Whieldon, D. Maturity sorting: New balance for young athletes. Physician Sp. Med. 6:127-132; 1978.
102. Whiteside, J.; Fleagle, S.; Kalanek, A.; Weber, H. Manpower loss in football: A 12-year study at the Pennsylvania State University. Physician Sp. Med. 13(1):103-114; 1985.
103. Wilson, K.; Rental, E.; Rental, M. Facial injuries in football. Am. Acad. Opthalm. Otolaryng. 77:434-437; 1973.
104. Zemper, E. Injury rates in a national sample of college football teams: A 2-year prospective study. Physician Sp. Med. 17(11):100-113; 1989a.
105. Zemper, E. A prospective study of injury rates in a national sample of American college football players. In: Proceedings of the First International Olympic Committee World Congress on Sport Sciences. Colorado Springs: USOC; 1989b:194-195.
106. Zemper, E. Cerebral concussion rates in various brands of football helmets. Athletic Training. 24(2):133-137; 1989c.
107. Zemper, E. A two-year prospective study of prophylactic knee braces in a national sample of college football players. Sports Training, Medicine and Rehabilitation. 1:287-296; 1990.
108. Zemper, E. Epidemiology of athletic injuries. In: McKeag, D.; Hough, D.; Zemper, E. Primary care sports medicine. Dubuque, IA: Brown & Benchmark;1993:63-73.
109. Zemper, E. Analysis of cerebral concussion frequency with the most commonly used models of football helmets. J. Athl. Training. 29(1):44-50; 1994.}}}}}}}

		High School	___________	___College____	________
		Alles et al., 1979	Alles et al., 1979	Buckley & Powell, 1979	Zemper, 1989b
	# injuries:	1696	12,432	7190	3744
	Study type:	P^a	P	P	P
Injury location____	# participants:	____2674____	___13,416___	___28,419___	____8325___
Head		0.62	0.78	1.7	0.44
Spine/trunk		0.63	0.49	0.9	0.87
Neck/spine		--	--	--	0.32
Trunk (torso)		0.63	0.49	0.9	--
Back		--	--	--	0.27
Chest/rib		--	--	--	0.15
Abdomen		--	--	--	0.02
Groin		--	--	--	0.11
Upper extremities		1.59	1.75	2.5	1.32
Shoulder		0.67	1.07	1.4	0.73
Arm (upper&lower)		0.92	0.68	1.1	0.05
Elbow		--	--	--	0.12
Wrist		--	--	--	0.05
Hand/finger		--	--	--	0.37
Lower extremities		4.61	6.20	8.1	3.34
Pelvis/hip		--	--	--	0.18
Knee		1.95	3.16	3.1	1.19
Leg (upper&lower)		1.19	1.30	2.4	0.85
Ankle		1.47	1.74	2.6	0.93
Foot/toe		--	--	--	0.19
Other		0.40	0.00	0.2	0.08

	_________	High School	_________	________	___College__	_______
Year_______	Direct __fatalities__	Indirect __fatalities__	Incomplete _recovery_	Direct __fatalities__	Indirect __fatalities__	Incomplete _recovery_
1970	23	12	--	3	2	--
1971	15	7	--	3	2	--
1972	16	10	--	2	1	--
1973	7	5	--	0	3	--
1974	10	5	--	1	3	--
1975	13	3	--	1	3	--
1976	15	7	--	--	2	--
1977	8	6	10	1	0	2
1978	9	8	13	0	1	0
1979	3	8	8	1	1	3
1980	9	4	11	0	0	2
1981	5	6	6	2	0	2
1982	7	7	7	0	3	2
1983	4	6	11	0	3	1
1984	4	3	5	1	0	0
1985	4	1	6	1	1	3
1986	11	7	3	1	1	0
1987	4	4	9	0	3	0
1988	7	10	10	0	0	1
1989	4	9	12	0	2	2
1990	0	3	11	0	3	2
1991	3	3	1	0	1	0
1992	2	8	4	0	1	0
Totals	183	142	127	17	36	20

	______High School_____		_____College_____
Year____	Fatality incidence	Catastrophic incidence	Fatality incidence	Catastrophic incidence
1970	1.92	--	4.00	0.00
1971	1.25	--	4.00	0.00
1972	1.33	--	2.67	--
1973	0.58	--	0.00	--
1974	8.83	--	1.33	--
1975	1.08	--	1.33	--
1976	1.00	--	0.00	--
1977	0.53	0.77	1.33	2.67
1978	0.60	1.00	0.00	0.00
1979	0.23	0.62	1.33	4.00
1980	0.69	0.85	0.00	2.67
1981	0.38	0.46	2.67	2.67
1982	0.54	0.54	0.00	2.67
1983	0.30	0.85	0.00	1.33
1984	0.30	0.38	1.33	0.00
1985	0.30	0.46	1.33	1.00
1986	0.77	0.23	1.33	0.00
1987	0.30	0.69	0.00	0.00
1988	0.46	0.77	0.00	1.33
1989	0.27	0.80	0.00	2.66
1990	0.00	0.73	0.00	2.66
1991	0.20	0.07	0.00	0.00
1992	0.07	0.27	0.00	0.00

		Culpepper, & Nieman, 1983	Blyth & Mueller, 1974	Olson, 1979	Prager, 1989		NATA, 1987^a
	Type of study:	Case series	Prospective	Prospective	Prospective		Prospective
Player	Type of rate:	% of injuries	% of injuries	% of injuries	% of injuries	% of injuries	Inj./100 games
_position_	# injuries:__	1877_____	4287_____	465______	251______	__range__	4292_____
Offense
End		--	--	12.8	--	-- - 12.8	3.4^b
Tackle		11.6	5.7	8.2	19.5^c	5.7-9.5	--
Guard		7.8	7.1	1.9	9.6^c	1.9-9.6	--
Center		4.3	3.6	2.3	5.2^c	2.3-5.2	--
Quarterback		8.7	6.1	3.9	6.4	3.9-8.7	6.9
Running back		19.6	21.1	26.1	15.9	15.9-26.1	9.3
Fl./wide recvr		7.2	3.3	--	2.0	2.0-7.2	2.9
Tight end		5.2	4.7	--	6.4	4.7-6.4	4.2
Totals		64.4	51.6	55.2	65.0	51.6-65.0	26.7
Defense
End		4.7	9.2	8.2	--	4.7-9.2	--
Lineman		5.5	16.7	17.9	--	5.5-17.9	4.7
Def. back		9.2	10.6	16.3^d	7.2	7.2-16.3	3.4
Linebacker		9.0	11.4	--	13.1	9.0-13.1	5.1
Half/cornerback		--	--	--	--	--	--
Safety		--	--	1.6	2.0	1.6-2.0	--
Totals		28.4	47.9	44.0	22.3	22.3-47.9	13.2
Kicker/punter		--	--	0.8	1.6	0.8-1.6	--
Other		7.2	0.2	--	11.1	0.2-11.1	--
Totals		7.2	0.2	0.8	12.7	0.2-12.7	--

		Canale, 1981	Whiteside, 1985	NCAA ISS, 1984	NCAA ISS, 1985	Zemper, 1989b
	Type of study:	Prospective	Prospective	Prospective	Prospective	Prospective
Player	Type of rate:	% of injuries	% of injuries	Per 1,000 A-E	Per 1,000 A-E	Per 1,000 A-E
_position_	# injuries:__	283______	2186_____	3218_____	3002_____	3744_____
Offense
End		7.8^a	28.9^a	0.81	0.81	0.30^b
Tackle		--	--	1.03	0.52	0.22
Guard		--	--	1.13	0.57	0.23
Center		--	--	0.46	0.46	0.18
Quarterback		5.9	1.3	0.58	0.58	0.26
Running back		21.6	14.3^c	1.88	0.95	0.43
Slot/wingback		--	--	0.16	0.16	0.06
Fl./wide recvr		7.8	6.3	1.03	1.03	0.48
Tight end		3.9	--	--	--	--
Totals		47.0%	50.8%	7.08	5.08	3.02
Defense
End		11.8	--	--	--	--
Lineman		17.6	17.9	2.17	0.43	0.23
Def. back		21.6	14.0^d	^--	--	--
Linebacker		2.0	14.6	2.04	0.68	0.32
Half/cornerback		--	--	1.36	0.69	0.29
Safety		--	--	0.82	0.82	0.15
Totals		53.0%	46.5%	6.39	2.62	0.99
Special teams		--	--	0.39	0.39	--
Kicker/punter		--	2.7	0.11	0.11	--
Other		--	--	0.05	0.05	--
Totals		--	2.7%	0.55	0.55	--

____Study________________Results____________
Grace et al., 1988	Matched pairs; Injury based on time loss from practice; Injuries graded from mild to severe	580 H.S. players over a period of two seasons; 247 wore single-hinged braces; 83 wore double-hinged braces; 250 wore no braces	Significantly more knee injuries in the single-hinged braces group than in double-hinged and nonbraced groups; Foot and ankle injuries more common in braced players; More sever injuries to lower extremity in braced group; More injuries to the knee than other lower extremity areas, MCL tears were most significant knee injury
Hansen et al., 1985	Longitudinal; Number and type of knee surgeries were analyzed; Team medical records were used	University of Southern California (1980-1984)	11% injury rate in nonbraced group, 5% injury rate in braced group; <2% of braced players required surgery, 5% on nonbraced players required surgery; Braced players had one-quarter as many meniscectomies; 1 braced player had severe knee injury, 6 nonbraced players had severe knee injuries; 4% of braced offensive linemen were injured, 18% of nonbraced linemen were injured; 7% of braced defensive linemen were injured, 23% of nonbraced linemen were injured
Hewson et al., 1986	Longitudinal; Team rosters, treatment logs, NAIRS, medical records were analyzed; Injury rates based on number of exposures in games and practices	University of Arizona 1977-1980 no brace use 1981-1985 braces used	Number, type, and severity of knee injuries similar in both groups; Braced and nonbraced players had similar chances of being injured; Knee injury prevention not improved by brace use
Rovere et al., 1987	Longitudinal; Injury records at University Sports Medicine Unit were analyzed; Injury rates based on incidence per 100 players	Wake Forest University 1981-1985 fall seasons and spring practices	More injuries, particularly Grade I MCL sprains, during the braced period; More knee operations performed during the braced period; Regardless of brace use, offensive team suffered more injuries; Knee injuries caused by body contact were fewer during braced period; More injuries occurred during games, with higher rates seen during the braced period
Schriner, 1985	Survey; Based on rates per 100 players; Time-loss parameters used; Injuries classified by injury mechanism involved	1,246 players from Michigan high schools; 1984 season; 197 players wore braces	Knee braces reduced injuries to the medial structures of the knee; No MCL, medial meniscus, and ACL injuries from lateral forces seen in braced players; Braced players did incur hyperextension knee injuries
Shaw and Brubaker, 1987	Longitudinal; Criteria for bracing and injury not defined	Texas City High School varsity and junior varsity; 1983-1986 seasons; 1983 no braces used; 1984 all varsity players wore braces; 1985-86 all players fitted with braces, but not required to wear them	In 1983, 19 knee injuries with 143 days of time loss; 1984 - 17 injuries, 5 days time loss for braced players, 90 days for nonbraced; 1985 - 9 injuries, 30 days time loss for braced players, 22 days for nonbraced; No associated structural damage to players wearing braces
Teitz et al., 1987	Survey; criteria for bracing, exposure while braced, and injury definition not stated	71 NCAA Division I teams in 1984; 61 NCAA Division I teams in 1985; 7,010 players braced, 4,584 players not braced	Braced players had higher injury rates, particularly running backs and defensive backs; MCL injuries were most common; Braced players missed as much or more playing time than nonbraced players; No difference in injury severity between braced and nonbraced players
Zemper, 1990	2-year prospective cohort study; Injury rates based on athlete-exposures; Brace use criteria determined by each school; One day or more time loss needed for injury to be reported	1986 season: 32 NCAA and NAIA teams, 3431 players; 1987 season: 27 NCAA and NAIA teams, 2798 players; 28% of players wore braces	Braces do not reduce MCL injury rate; More time loss for those wearing braces; Approximately half of reported knee injuries occurred in games; No significant difference in knee surgery rate for braced and nonbraced players; No difference between the different brands of braces used; No relationship between brace use and ankle injuries; More MCL injuries among braced players who played on artificial turf; No significant effect of braces in reducing the severity of knee injuries

____Preventive recommendation____	_________________Discussion and Support__________________
Preparticipation examination (Abdenour and Weir, 1986; Blackwell and McCullagh, 1990; Boynton, 1988; Brooks et al., 1976; Caine and Broekhoff, 1987; Coddington and Troxell, 1980; Lysens et al., 1989; Riley, 1978; Robey et al., 1971; Taylor and Luckstead, 1987)	A comprehensive medical history, physical and mental examination prior to athletic participation should be obtained and administered to all players. Abdenour and Weir (1986) demonstrated that historical information about the athlete was a significant predictor of physical well-being. The examination will then serve as a screening device to detect any preexisting medical abnormalities (cardiovascular, neurological, musculoskeletal, respiratory, etc.) and/or physical weaknesses. It will provide baseline information about the health and injury status of the player and will also aid in the detection of prior injury, which has been shown to increase the chance of further injury. The psychological status of the athlete should also be examined and monitored because such factors as life stress, risk-taking behavior, and competitive anxiety have been shown to increase the likelihood of injury.
Maturity assessment (Caine and Broekhoff, 1987; Robey et al., 1971; Whieldon, 1978)	It appears that the risk of injury is not only related to age, but also to the size, physical, and sexual development of the athlete. Caine and Broekhoff (1987) recommend that players be matched according to their level of maturation (size, strength, power, skill level, fitness, experience) to prevent mismatching unequally developed players, thereby preventing unsafe and unfair play.
Year-round injury and health evaluation	This would provide the athletic program with an effective tracking system to constantly evaluate and identify any predisposing conditions that may make the players more susceptible to further more serious injury.
Adequate conditioning (physical and mental) and training of athletes (Boynton, 1988; Brooks et al., 1976; Lysens et al., 1989; Riley, 1978)	A year-round (preseason, inseason, and postseason) mandatory football-specific conditioning and training program should be aimed at improving muscular and ligament imbalances and weaknesses; coordination and timing; flexibility, mobility, and agility; and cardiovascular and endurance capacities. Cahill (1977) reported that players on a weight training program had a decreased incidence of knee injuries because of increased strength of the surrounding ligaments, muscles, and tendons. The conditioning and training program should be a well-monitored gradual progression program with care taken not to overstress the athletes too soon. Physical contact should not be introduced until a general level of fitness has been established. Boynton (1988) states that conditioning has an added quality in that improvement in physical conditioning will lead to subsequent improvements in the mental attitude of the players and of the team as a whole (e.g., confidence and controlled aggression). An adequate and effective treatment and rehabilitation program should also be an integral part of the overall conditioning program of the athlete.
Limited contact activity (Blyth and Mueller, 1974; Cahill and Griffith, 1979; Robey et al., 1971)	Live contact between players can be limited by providing field equipment to teach and practice certain activities. However, safe techniques need to be established when using this equipment as several studies have shown a relationship between field equipment and the incidence of injury. When comparing injury incidence between limited and regular contact programs, Blyth and Mueller (1974) reported there were fewer injuries in the limited contact program (13.3%) than the regular contact program (86.5%).
Adequate provision of medical attention and treatment for the athletes (Blyth and Mueller, 1974; Boynton, 1988; Lackland et al., 1982; Truxal et al., 1981)	The early and effective diagnosis, evaluation, treatment, monitoring, and subsequent prevention of injuries may be accomplished by adequate on-field medical coverage (physicians, sports medicine personnel, and athletics trainers) at all games and practices. Blyth and Mueller (1974) reported that 2.4% of the injuries were the result of inadequate medical care and extreme heat, with only half of the coaches providing liquid to their players during activity. Lackland et al. (1982) demonstrated the value of athletic trainer/team physician services in their high school football injury study. Proper medical supervision during football participation will ensure that activities are performed under safe playing conditions (e.g., avoiding extreme heat, humidity, and cold).
Education, training, and certification of coaches (Caine and Broekhoff, 1987; Blyth and Mueller, 1974; Boynton, 1988)	All coaches should have adequate coaching and playing experience and be trained and evaluated in the fundamentals of the game. The use and coaching of illegal, unsafe techniques should not be tolerated. Coaches should be required to take educational and certification sport and exercise-related courses/clinics to improve, maintain, and refresh their knowledge of the game. A capable, well-trained assistant coaching staff should be a part of all football programs. Schools with several assistant coaches were shown to have a lower incidence rate than those schools without assistant coaches (Blyth and Mueller, 1974).
Rules enforcement (Amold and Coker, 1977; Blyth and Mueller, 1974; Boynton, 1988)	Coaches and athletes need to abide by the rules of the game, with referees and officials playing an active role in their enforcement. The 3.8% of injuries caused by illegal acts such as clipping, spearing, hitting while down and out-of-bounds, and personal fouls reported in the study by Blyth and Mueller (1974) may have been eliminated with proper rule awareness and enforcement.
Equipment (Blyth and Mueller, 1974; Boynton, 1988; Garrick et al., 1978; Johnson and Ritter, 1975; Schwank and Ryan, 1975; Slagle, 1983).	The use of proven protective equipment needs to be an integral part of all football programs. Additional padding and/or taping for the ribs, thighs, arms, hands, and ankles may help prevent some injuries. Additionally, the management of properly fitting and well-maintained equipment needs to be stressed. In the study directed by Blyth and Mueller (1974) 1.7% of all injuries were caused by ill-fitting, defective, or broken equipment. The fields and playing surfaces need to be inspected and well-maintained. Garrick et al. (1978) recommends that a buffer zone be placed around the periphery of the playing field to prevent injuries occurring as the result of contact with people or objects out-of-bounds. Coaches and players should be prohibited from encouraging the use of protective equipment as offensive weapons. Further research needs to be directed at improving the protective equipment being used by football players.