Article, Sports Medicine

Tackling causes and costs of ED presentation for American football injuries: a population-level study

a b s t r a c t

Background: American tackle football is the most popular high-energy impact sport in the United States, with ap- proximately 9 million participants competing annually. Previous epidemiologic studies of football-related inju- ries have generally focused on specific geographic areas or pediatric age groups. Our study sought to examine patient characteristics and outcomes, including hospital charges, among athletes presenting for emergency de- partment (ED) treatment of football-related injury across all age groups in a large nationally representative data set.

Methods: Patients presenting for ED treatment of injuries sustained playing American tackle football (identified using International Classification of Diseases, Ninth Revision, Clinical Modification code E007.0) from 2010 to 2011 were studied in the Nationwide Emergency Department Sample. Patient-specific injuries were identified using the primary International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code and categorized by type and anatomical region. Standard descriptive methods examined patient demographics, diagnosis categories, and ED and inpatient outcomes and charges.

Results: During the study period 397 363 football players presented for ED treatment, 95.8% of whom were male. Sprains/strains (25.6%), limb fractures (20.7%), and head injuries (including traumatic brain injury; 17.5%) repre- sented the most presenting injuries. Overall, 97.9% of patients underwent routine ED discharge with 1.1% admit- ted directly and fewer than 11 patients in the 2-year study period dying prior to discharge. The proportion of admitted patients who required surgical interventions was 15.7%, of which 89.9% were orthopedic, 4.7% neuro- logic, and 2.6% abdominal. Among individuals admitted to inpatient care, mean hospital length of stay was

2.4 days (95% confidence interval, 2.2-2.6) and 95.6% underwent routine Discharge home. The mean total charge for all patients was $1941 (95% confidence interval, $1890-$1992) with substantial injury type-specific variabil- ity. Overall, at the US population, estimated Total charges of $771 299 862 were incurred over the 2-year period. Conclusion: In this nationally Representative sample, most ED-treated injuries associated with football were not acutely life threatening and very few required major therapeutic intervention. This study provides a cross- sectional overview of ED presentation for acute football-related injury across age groups at the population level in recent years. Longitudinal studies may be warranted to examine associations between the patterns of in- jury observed in this study and long-term outcomes among American tackle football players.

(C) 2016

? All authors report no potential conflicts of interest.

?? All authors have read and approved the submitted manuscript; the manuscript is not under consideration for publication elsewhere, nor has it been published elsewhere in whole or in part, except as an abstract.

* Corresponding author at: Center for Surgery and Public Health, Brigham and Women’s Hospital, Harvard Medical School, 1620 Tremont St, Suite 4-020Y, Boston, MA. Tel.: +1 617 525 6696.

E-mail address: [email protected] (E.B. Schneider).

Introduction

American tackle football is the most popular high-energy impact sport in the United States, with approximately 9 million participants competing annually [1,2]. Tackle football is associated with the highest injury rates among all organized Team sports, with previous studies reporting between 300 000 and 1.2 million football-related injuries

http://dx.doi.org/10.1016/j.ajem.2016.02.057

0735-6757/(C) 2016

annually among high school football players alone [3-9]. Recent reports in the popular media have focused on football-related head injuries, in- cluding acute disorders such as concussions and the possible chronic ef- fects of repeated head injury, which has been described as chronic traumatic encephalopathy [10-12]. In addition, football-related head trauma has been associated with increased long-term risks of depres- sion, substance abuse, cognitive dysfunction, and, possibly, dementia [13-20].

Although head trauma is the subject of much study in football, there are few studies that have examined the broad incidence of football- related injuries of all types among football players of different ages. Prior epidemiologic studies of football-related injury generally have been restricted to specific geographic areas, or to players/patients in pe- diatric age groups [6,21-24]. To date, there is limited published infor- mation quantifying hospital resource utilization and hospital charges associated with the entire range of football-related injuries.

We sought to examine factors associated with emergency depart- ment (ED) presentation and patient outcomes, including hospital charges, among patients seeking ED treatment of football-related inju- ries across all age groups in a large nationally representative data set.

Methods

The Nationwide Emergency Department Sample (NEDS) is the larg- est all payer ED database in the United States, which can be weighted to provide estimates of hospital-based ED visits at the level of the US na- tional population. In 2011, the NEDS provided discharge data from 950 hospitals located throughout 30 US states, representing a 20% strat- ified sample of ED visits in the United States. At the population level, ap- proximately 130 million ED visits are represented in the NEDS annually [25]. Patient-specific external cause of injury codes (E-codes) were used to identify the subset of individuals who presented with football-related injuries. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) was modified in 2009 to include a code for external cause of injury specific to American tackle football (E-code: 007.0, “Activities involving tackle football played as a team or group.”), which was used to identify subjects in this study. This study was approved by the Institutional Review Board of the Johns Hopkins Medical Institutions.

Study sample

All patients presenting for ED treatment of injuries sustained playing tackle football in 2010 and 2011 were selected for study using the football-specific E-code. For each football-related ED presentation, the principal diagnosis was identified using the primary ICD-9 diagnosis code and categorized as follows: (1) sprains and strains of joints and ad- jacent muscles; (2) injury to the head; (3) injury to the spinal cord;

(4) injury to nerves and plexuses; (5) limb fractures; (6) contusions and superficial injuries; (7) crushing injury; (8) open wounds; (9) dislo- cations; (10) internal injuries of chest, abdomen and pelvis; (11) trau- matic complications; and (12) other injuries (including injuries to blood vessels and late effects of musculoskeletal injuries and burns of limbs, wrist, and hands). It is important to note that these injury- related groupings do not represent entirely homogeneous subsets of pa- tients who might be expected to experience similar treatment and out- comes. For example, injuries specific to the head included the following spectrum of diagnoses: intracranial injuries, head injuries not otherwise specified, abrasions and contusions of the face/scalp, open wounds of the head, and Skull fractures (Supplemental Table 1).

In addition to the primary ICD-9 diagnosis code, each ED presenta- tion record includes up to a maximum 14 secondary Diagnosis codes. These secondary diagnosis codes were examined to determine if they were related to the principal injury (eg, was a patient with a primary di- agnosis of limb fracture likely carry a secondary diagnosis of strain/

sprain, or were secondary diagnoses likely to represent substantial inju- ries to other body regions).

Patient age at presentation was examined as a categorical variable as follows: children younger than 11 years, adolescents 11 to 14 years old, older adolescents aged 15 to 18 years, young adults aged 19 to 22 years, and older adults 23 years or older. The ranges for these 5 age groupings were selected as they approximately represent elementary, middle school, high school, college, and professional/semiprofessional/adult recreational levels, respectively. Patient demographics, including sex, insurance status, zip code-based Household income quartile, month of visit, location of visit (urban vs rural), and hospital teaching status were examined and compared across the 5 different age groups using standard descriptive statistical methods. To understand age-related var- iability in injury type, the proportions of patients with primary diagno- ses in each of the 12 injury categories were compared across age groups. Patient-level outcomes of interest included ED discharge disposition (eg, discharge to home, transfer, or admission to inpatient status, etc). Among patients admitted to inpatient care, age-related proportional differences in the incidence of major Therapeutic procedures (identified using ICD-9 procedure codes and grouped into 4 categories as follows: orthopedic, neurologic, vascular, or general [Supplemental Table 2]) were examined. In addition, among patients admitted to inpatient care, hospital discharge disposition was examined across age groups. Total hos- pital charges were computed and compared across age groups and diag- nostic categories. Patients with missing data such that they could not be categorized were excluded from analysis. Population-specific rates were calculated using intercensal estimates from the US Census Bureau. Data were weighted to represent the national population and all analyses

were conducted using Stata 12.1 MP (College Station, TX).

Results

A total of 397 363 patients of all ages met the study criteria in 2010 and 2011. Overall, most patients were male (380 493 [95.8%]), with the 15- to 18-year age group containing the smallest proportion of females (2.7%). Most patients had private insurance (208 383 [52.4%]). Howev- er, 35.3% of patients 18 years or younger had primary coverage through Medicaid compared with 13.6% of patients 19 years or older (P b .001). The distribution of zip code-based household income quartile did not vary within any age group or between age groups. Across patients of all ages, most presented for treatment on a weekday (277 967 [70.0%]

); however, patients 19 years and older were proportionally more likely to present on a weekend compared with patients 18 years and younger (42.9% vs 27.4%, P b .001). Most patients presented during the months of August through November, which encompass the traditional football season. Age-related patterns appeared for month of admission, with pa- tients 19 years and older presenting for ED treatment more uniformly throughout the year, whereas patients 18 years and younger were pro- portionally more likely to present during the months of August through November. Most patients presented for ED treatment in urban locations (83.5%) and in nonteaching hospitals (65.0%). Population-specific rates of injury were 21.3 per 100 000 in the Midwest, 17.1 in the West, 15.8 in the South, and 10.5 in the Northeast (Table 1).

More than one quarter of all patients presented with a primary diag- nosis of sprains and strains of joints and adjacent muscles (25.6%). One (20.7%) in 5 patients presented with a primary diagnosis of limb frac- ture, of which the majority (78.3%) involved an upper limb. The next most common diagnosis grouping included injuries to the head (69 483 [17.5%]). The proportion of patients with a primary diagnosis related to head injury was greatest in the 15- to 18-year age group (20.3%) and smallest among patients 23 years or older (10.6%). Within injury categories, the proportional diagnosis of specific injury subtypes demonstrated substantial variability. For example, among patients of all ages classified as having head injury, the most commonly reported specific diagnosis was Intracranial injury without fracture (7.3%) follow- ed by head injury not otherwise specified (4.3%) and open wound to the

Table 1

Patient characteristics across age groups

<= 10 y

11-14 y

15-18 y

19-22 y

>= 23 y

Total

(n = 47 914), n

(n = 150 969), n

(n = 131 459), n

(n = 25 255), n

(n = 41 767), n

(n = 397 363), n

(%)

(%)

(%)

(%)

(%)

(%)

Sex

Male

45 653

(95.28)

145 668

(96.49)

127 839

(97.25)

23 732

(93.97)

37 603

(90.03)

380 493

(95.75)

Female

2261

(4.72)

5302

(3.51)

3620

(2.75)

1523

(6.03)

4164

(9.97)

16 870

(4.25)

Payer

Medicare

228

(0.48)

957

(0.63)

759

(0.58)

233

(0.92)

1412

(3.38)

3589

(0.90)

Medicaid

19 928

(41.59)

55 553

(36.80)

41 065

(31.24)

3516

(13.92)

5570

(13.34)

125 632

(31.62)

Private

23 381

(48.80)

79 333

(52.55)

74 635

(56.77)

12 426

(49.20)

18 609

(44.55)

208 383

(52.44)

Self-pay

2527

(5.27)

9023

(5.98)

9322

(7.09)

7209

(28.54)

12 903

(30.89)

40 983

(10.31)

Others

1851

(3.86)

6103

(4.04)

5679

(4.32)

1871

(7.41)

3273

(7.84)

18 776

(4.73)

Median household income

$1-$38 999

12 305

(25.68)

35 652

(23.62)

31 894

(24.26)

6653

(26.34)

10 169

(24.35)

96 673

(24.33)

$39 000-$47 999

11 647

(24.31)

37 279

(24.69)

33 240

(25.29)

6413

(25.39)

10 361

(24.81)

98 940

(24.90)

$48 000-$62 999

11 857

(24.75)

39 187

(25.96)

33 044

(25.14)

5688

(22.52)

10 685

(25.58)

100 461

(25.28)

$63 000 or more

11 528

(24.06)

36 751

(24.34)

31 224

(23.75)

6110

(24.19)

9663

(23.14)

95 275

(23.98)

Unreported

576

(1.20)

2101

(1.39)

2057

(1.56)

391

(1.55)

890

(2.13)

6015

(1.51)

Weekend admission

No

31 980

(66.74)

108 274

(71.72)

99 449

(75.65)

14 740

(58.36)

23 523

(56.32)

277 967

(69.95)

Yes

15 933

(33.25)

42 695

(28.28)

32 010

(24.35)

10 515

(41.64)

18 239

(43.67)

119 391

(30.05)

Admission month

January

925

(1.93)

2212

(1.47)

1509

(1.15)

959

(3.80)

1647

(3.94)

7252

(1.83)

February

911

(1.90)

1942

(1.29)

1724

(1.31)

901

(3.57)

1316

(3.15)

6795

(1.71)

March

1570

(3.28)

2819

(1.87)

2466

(1.88)

1070

(4.24)

1859

(4.45)

9784

(2.46)

April

1508

(3.15)

3828

(2.54)

2863

(2.18)

1704

(6.75)

2584

(6.19)

12 487

(3.14)

May

1328

(2.77)

4138

(2.74)

4312

(3.28)

1159

(4.59)

2754

(6.59)

13 691

(3.45)

June

866

(1.81)

2446

(1.62)

4268

(3.25)

1090

(4.32)

2515

(6.02)

11 186

(2.82)

July

1062

(2.22)

2799

(1.85)

4276

(3.25)

1193

(4.72)

2502

(5.99)

11 833

(2.98)

August

6462

(13.49)

23 975

(15.88)

19 789

(15.05)

1886

(7.47)

2948

(7.06)

55 059

(13.86)

September

10 841

(22.63)

40 304

(26.70)

32 160

(24.46)

3354

(13.28)

4691

(11.23)

91 349

(22.99)

October

10 284

(21.46)

34 358

(22.76)

29 247

(22.25)

3978

(15.75)

5310

(12.71)

83 177

(20.93)

November

3885

(8.11)

9736

(6.45)

9732

(7.40)

3147

(12.46)

5715

(13.68)

32 217

(8.11)

December

1201

(2.51)

2603

(1.72)

2056

(1.56)

1210

(4.79)

1903

(4.56)

8974

(2.26)

Unreported

7070

(14.76)

19 808

(13.12)

17 057

(12.98)

3603

(14.27)

6022

(14.42)

53 561

(13.48)

Teaching status

Not teaching

30 434

(63.52)

98 615

(65.32)

88 052

(66.98)

15 663

(62.02)

25 461

(60.96)

258 227

(64.99)

Teaching

17 479

(36.48)

52 354

(34.68)

43 407

(33.02)

9591

(37.98)

16 306

(39.04)

139 137

(35.02)

Hospital location

Rural

7156

(14.94)

26 030

(17.24)

24 306

(18.49)

3433

(13.59)

4485

(10.74)

65 410

(16.46)

Urban

40 757

(85.06)

124 939

(82.76)

107 152

(81.51)

21 822

(86.41)

37 282

(89.26)

331 953

(83.54)

Hospital region

North-East

4861

(10.15)

16 547

(10.96)

14 714

(11.19)

3825

(15.15)

6676

(15.98)

46 623

(11.73)

Midwest

13 268

(27.69)

44 442

(29.44)

38 535

(29.31)

6959

(27.55)

11 091

(26.55)

114 294

(28.76)

South

20 326

(42.42)

54 376

(36.02)

44 020

(33.49)

8303

(32.88)

13 416

(32.12)

140 441

(35.34)

West

9459

(19.74)

35 604

(23.58)

34 190

(26.01)

6167

(24.42)

10 585

(25.34)

96 005

(24.16)

head (3.4%). The mean head-specific Abbreviated Injury Severity score was lower for patients 18 years and younger than for those older than 18 years (1.43 [95% confidence interval {CI}, 1.41-1.44] vs 0.77 [95% CI, 0.73-0.81]). Fewer than 1000 patients presented across the entire study period with a primary diagnosis of internal injury (0.25%), and only 0.09% of patients presented with a primary diagnosis related to in- juries to the spinal cord (Table 2). Not surprisingly, among patients pre- senting with multiple injuries, the nonprimary injures were generally related to the patient’s principal diagnosis. Almost all patients present- ing for treatment of football-related injuries were discharged directly to home from the ED (388 373 [97.7%]) with little variation observed across age groups. A total of 4598 (1.2%) were admitted to inpatient sta- tus at the index hospital, and 3354 (1.0%) patients underwent direct transfer from the presenting ED to another facility (Table 3).

Patients with internal injuries and injuries to the spinal cord were proportionally most likely to be admitted to inpatient care, 44.5% and 46.5% respectively. Among the 4589 patients admitted to the hospi- tal, 722 (15.7%) underwent a major therapeutic procedure. The pro- portion of all surgical interventions that were orthopedic in nature was 89.9%, half of which involved the lower extremities (Fig. 1). Among individuals admitted to inpatient care, mean hospital length of stay was 2.4 days (95% CI, 2.2-2.6). Patients admitted with a pri- mary diagnosis of internal injury of the chest, abdomen, or pelvis

demonstrated the longest injury type-specific inpatient mean length of stay (3.4 days [95% CI, 2.9-3.8]). A dose-response relationship be- tween age and length of stay was observed, with mean length of stay being shortest among the youngest patients at 1.7 days (95% CI, 1.5- 1.9) and increasing progressively to 3.0 days (95% CI, 2.4-3.6) in the oldest age group. As was the case for patients treated and released from the ED, most patients admitted to inpatient care ultimately underwent routine discharge to home (95.6%). Across the 2-year study period, fewer than 11 patients presenting for ED treatment of football-related injuries died prior to discharge.

The mean total charge for all patients presenting for ED treatment of football-related injury was $1941 (95% CI, $1890-$1992), with the mean total charges being highest among patients in the 15- to 18-year age group ($2162 [95% CI, $2101-$2223]). Substantial injury type- specific variability in mean total charges was observed with patients presenting for treatment of football-related injury. For example, pa- tients presenting with a primary diagnosis of spinal cord injury incurred average total charges of $17 028 (95% CI, $11 969-$22 087), whereas total charges associated with sprains and strains averaged $1211 (95% CI, $1157-$1265). Most injuries fell within the category encompassing sprains and strains; however, the percent total charges associated with these injuries was less than percent total charges associated with fractures or head injuries (Fig. 2). Overall, at the level of the US

Table 2

Diagnoses across age groups

Diagnosis category

<= 10 y

(n = 47 914), n (%)

11-14 y

(n = 150 969), n (%)

15-18 y

(n = 131 459), n (%)

19-22 y

(n = 25 255), n (%)

>= 23 y

(n = 41 767), n (%)

Total

(n = 397 363), n (%)

Sprains and strains of joints and adjacent muscles

10 488

(21.89)

36 967

(24.49)

34 001

(25.86)

7252

(28.72)

13 086

(31.33)

101 795

(25.62)

Injury to the head

9360

(19.5)

24 012

(15.9)

27 333

(20.8)

4361

(17.3)

4417

(10.6)

69 483

(17.5)

  • Intracranial injury, without skull fracturea

2505

(5.2)

10 973

(7.3)

13 450

(10.2)

1062

(4.2)

855

(2.1)

28 845

(7.3)

  • Head injury not otherwise specified

2702

(5.6)

7122

(4.7)

6309

(4.8)

580

(2.3)

513

(1.2)

17 226

(4.3)

  • Open wounds of head

2489

(5.2)

2671

(1.8)

4433

(3.4)

1705

(6.8)

2052

(4.9)

13 350

(3.4)

  • Fracture of the skull

133

(0.3)

457

(0.3)

1027

(0.8)

619

(2.5)

543

(1.3)

2779

(0.7)

  • Abrasions and contusions of face and scalp

1531

(3.2)

2789

(1.9)

2114

(1.6)

395

(1.6)

454

(1.0)

7283

(1.9)

Injury to spinal cord

42

(0.09)

109

(0.07)

137

0.10

28

0.11

27

0.06

185

0.09

Injury to nerves and plexuses

22

(0.05)

47

(0.03)

52

(0.04)

14

(0.06)

22

(0.05)

157

(0.04)

Limb fractures

11 804

(24.63)

38 610

(25.58)

21 314

(16.21)

3832

(15.17)

6599

(15.80)

82 159

(20.67)

  • Fracture of upper limb

10 111

(21.10)

31 306

(20.74)

15 685

(11.93)

2620

(10.37)

4711

(11.28)

64 432

(16.21)

  • Fracture of lower limb

1693

(3.53)

7304

(4.84)

5629

(4.28)

1212

(4.80)

1888

(4.52)

17 727

(4.46)

Contusion and superficial injury

7487

(15.62)

23 003

(15.23)

17 353

(13.20)

2624

(10.39)

5478

(13.12)

55 944

(14.08)

Crushing injury

53

(0.11)

211

(0.14)

192

(0.15)

19

(0.08)

42

(0.10)

517

(0.13)

Open wounds

922

(1.92)

2076

(1.38)

2414

(1.84)

572

(2.26)

733

(1.76)

6718

(1.69)

Dislocations

747

(1.56)

3483

(2.31)

7732

(5.88)

2399

(9.50)

3478

(8.33)

17 838

(4.49)

Internal injury of chest, abdomen, and pelvis

49

(0.10)

237

(0.16)

521

(0.40)

79

(0.31)

106

(0.25)

992

(0.25)

Traumatic complications and unspecified injuriesb

3250

(6.78)

8800

(5.83)

7056

(5.37)

1334

(5.28)

2313

(5.54)

22 754

(5.73)

Othersc

3711

(7.75)

13 460

(8.92)

13 406

(10.20)

2756

(10.91)

5488

(13.14)

38 821

(9.77)

a Includes concussion, cerebral contusion, and intracranial hemorrhage.

b Air or Fat embolism as an early complication of trauma, traumatic shock, traumatic compartment syndrome, other injury of trunk, and so on.

c Injuries to blood vessels, late effects of musculoskeletal, late effects and superficial injuries, burns of upper limb, lower limb, writs and hands, and so on.

population, estimated total charges of $771 299 862 were incurred by patients presenting for ED treatment of football-related injuries across the country from 2010 to 2011.

Discussion

American tackle football is a substantial contributor to injury burden in the ED, regardless of age or injury type. Because of the violent nature of the sport, football is known to be associated with injury to players; however, little has been reported regarding the incidence of, and out- comes associated with acute injury of all types among football players of all ages. This current study provides a foundational description of the epidemiology of injuries leading to ED presentation that occur among football players of all ages at the level of the US population.

Although the criteria used to select our sample did not change by age group, the proportion of patients in each age group that take part in or- ganized, league-based tackle football diminish substantially as age in- creases. Recent data demonstrate that there are approximately 1.1 million, 71 300, and 2000 high school, National Collegiate Athletic Asso- ciation (NCAA), and National Football League (NFL) football partici- pants, respectively [26]. Players on NCAA and NFL teams have access to athletic trainers, team-affiliated physicians, and other forms of non- ED urgent care. Given the much larger number of individuals playing on organized high school football teams, it is likely that the under 18 years of age cohort contains more participants engaged in formal, or- ganized tackle football compared with older age groups. Furthermore,

given that the greater access to non-ED urgent care enjoyed by colle- giate and professional participants, they may be less likely to present for treatment in an ED captured by the NEDS relative to those of a sim- ilar age who are engaging in football activities informally. These infor- mal football activities are also more likely to result in lower-speed and lower-energy impact injuries than those of collegiate and professional players. Therefore, our observations relating to injury type and severity among college-aged and professional-aged populations are likely more reflective of the characteristics of these non-professional, noncollegiate “weekend warriors” rather than their higher-level counterparts.

Most players observed in this study were male, which is in accord with reports from previous studies [1,24,27-29]. Population-specific in- jury rates were more than twice as high in the Midwest than in the Northeast [30]. Interestingly, most primary injuries leading to ED pre- sentation were orthopedic in nature, which is in concordance with ear- lier reports [6,29,31]. However, nearly 1 in 5 football-related injuries associated with ED presentation involved injury to the head. Across all ages, relatively few individuals experienced life-threatening trauma, such as injury to the spinal cord or internal organs. As expected, most patients of all ages underwent routine discharge from the ED, a finding that parallels an earlier report on ED discharge dispositions among 6- to 17-year-old football players. Only 1.2% of patients presenting for ED treatment of football-related injuries were admitted to inpatient care at the index hospital, and fewer than 0.2% of ED patients ultimately underwent an emergency surgical procedure, 90% of which were ortho- pedic in nature. Not surprisingly, patients with injuries to the spinal

Table 3

ED Disposition across age groups

Disposition from the ED

<= 10 y

(n = 47 914), n (%)

11-14 y

(n = 150 969), n (%)

15-18 y

(n = 131 459), n (%)

19-22 y

(n = 25 255), n (%)

>= 23 y

(n = 41 767), n (%)

Total

(n = 397 363), n (%)

Routine discharge

46 926

(97.94)

147 962

(98.01)

128 140

(97.48)

24 656

(97.63)

40 690

(97.42)

388 373

(97.74)

Transfer to short-term hospital

452

(0.94)

1215

(0.80)

905

(0.69)

95

(0.38)

212

(0.51)

2879

(0.72)

Other transfer*

60

(0.13)

222

(0.15)

124

(0.09)

22

(0.09)

48

(0.11)

475

(0.12)

Home health care

NR

(0.01)

29

(0.02)

46

(0.03)

0

0.00

17

(0.04)

96

(0.02)

Against medical advice

77

(0.16)

255

(0.17)

143

(0.11)

128

(0.51)

197

(0.47)

801

(0.20)

Inpatient Admission

389

(0.81)

1249

(0.83)

2044

(1.55)

350

(1.39)

566

(1.36)

4598

(1.16)

Death

0

0.00

0

0.00

NR

(0.00)

0

0.00

NR

0.00

NR

(0.00)

Not admitted (unknown)

NR

(0.01)

37

(0.02)

53

(0.04)

NR

(0.02)

37

(0.09)

137

(0.03)

NR, not reportable, value <=11.

* Skilled nursing facility, intermediate care, and other types of facility.

Fig. 1. Major therapeutic procedures.

cord, nerves, and internal organs were substantially more likely to be admitted to inpatient care than those with other injury types.

Head injuries were the third most common diagnosis associated with ED presentation for football-related injury, and head injury was proportionally more common among high school-aged athletes com- pared with college-aged athletes, which parallels the findings of earlier reports [6,29,31]. Moreover, among patients presenting for ED treat- ment, head injury was proportionally least common among athletes 23 years and older, and mean head-specific Abbreviated injury severity scores were significantly smaller in patients 19 years and older. Al- though these findings conflict with those of Gessel et al [31], our find- ings support the notion that the informal weekend warriors sustain lower-speed and lower-energy impact injuries. The small number of deaths observed in our study (b 11) correlates with the findings of a Na- tional Center for Catastrophic Sport Injury Research report for the same study period (2010-2011), which reported 9 deaths among injured football players during the 2-year study period [32].

In general, the findings from this present study are similar to those reported in previous studies [22,23]. For example Nation et al [22] re- ported that 31.3% of football players 6 to 17 years old who were treated in an ED between 1990 and 2007 had a primary diagnosis of sprain/ strain injury. However, the overall number of patients presenting for ED treatment of football-related injuries at the national level may be somewhat lower than the numbers suggested in previous reports. Ear- lier studies have often relied on data from the National Electronic Injury surveillance System (NEISS), with many of these studies focused on a specific injury type or on the spectrum of injuries in certain age groups [21-24]. Studies using the NEISS have estimated that 220 000 to 290 000 tackle football-related ED visits occur annually among children 6 to 17 years old [21,22]. Using data from the NEDS, we identified ap- proximately 199 000 annual ED visits for patients of all ages, which sug- gests that our estimates may be quite a bit lower than those obtained using the NEISS. It should be noted that the data collection methods that underlie the NEISS are substantially different from those used to

Fig. 2. Presentation frequency and total hospital charge.

generate the NEDS data sets. Although the NEDS collects data on pa- tients who visit the ED for any reason, the NEISS was designed and im- plemented to collect data specifically related to injuries as they present in the ED. It is possible that database-specific differences in sampling methodology and data collection/abstraction protocols between the NEISS and NEDS might underlie the observed differences in population-level estimates of the annual incidence of ED presentation for football-related injury. For example, the NEISS is focused on injury- related ED presentations, it relies on a sample of 66 hospitals and 500 000 cases to generate national-level estimates, whereas the NEDS generates national-level estimates using approximately 30 000 000 all-cause ED visits using data provided by 950 hospitals, but relies on ad- ministrative billing records to identify diagnoses, procedures, and exter- nal causes of injury [25]. It is possible that the incidence of ED presentation for football-related injury may be underestimated in the NEDS; because it is unlikely that identifying football as a mechanism of injury would result in altered billing, it is possible that sport- specific ICD-9 E-codes may not be reported for all injuries related to tackle football. Although it is possible that our use of the NEDS might have resulted in an underestimate of the actual number of injuries, it is not clear that any such underestimate would be differential across age groups or injury types; with the caveat that football as a mechanism of injury may be systematically underreported in ICD-9-CM-based NEDS descriptors, our study provides national estimates on demograph- ic information of athletes across all age groups.

Although the game of football is associated with the highest player- specific rate of severe sports-related injuries [33], only a very small pro- portion of injuries require hospitalization with an even smaller portion of patients undergoing a major surgical procedure. Not surprisingly nearly 90% of all emergent/urgent surgical procedures (those occurring within the index hospitalization) were orthopedic. Weighted to the level of the US population, football-related injury treatment charges amounted to a total of $771 299 862 over the 2-year study period. If the NEDS underreports football-related injuries and the NEISS-based studies provide a more accurate overall estimate of ED visits for football-related injuries, then the annual charge estimates for football- related injuries at the population level would be markedly larger than those calculated using the NEDS. Regardless of which estimate is more accurate, the US Financial burden of football-related injuries in the hos- pital setting is substantial.

If public health interventions are to be successful at reducing the burden of football-related injury in the ED, they must be informed by relevant evidence. These data suggest that the population-level burden of football-related injury varies with age. Given the nearly 95% reduc- tion in the population of those involved in organized football after age 18, it is likely that the largest public health gains regarding prevention of football-related injuries will come from focusing on participants en- gaged in organized football at the high-school level. The population- level burden of football-related injuries for those older than 18 years is primarily due to play by nonprofessional, noncollegiate athletes. Thus, although the NCAA and NFL are right to take focused measures to reduce the burden of injuries in their focused populations, the general population may be better served by including questions on hobbies and recreation in formal or informal social histories taken in the examina- tion room, which may provide an opportunity for discussion on strate- gies for injury prevention.

Although the findings of this study may not alter individual-level care for injured football players, emergency medicine physicians have the opportunity to provide decision makers in the community with a unique perspective on injury and injury prevention. From advocacy for prevention to treatment policy development and implementation, having accurate information on population-level injuries enables emer- gency physicians to impact health care in at-risk populations. The role of advocacy and policy development are particularly important at this time considering the ongoing shift of Hospital reimbursement toward payment bundles, which are often calculated using assumptions

derived from population-level health care data. By understanding both population-level injury data and patient-specific injuries and treat- ments, emergency physicians can contribute greatly to maintaining availability of best-practice care, even for Young athletes injured playing football.

As mentioned above, a major limitation of this study is that the NEDS is an administrative database and does not have more comprehensive sports-specific information for the patients admitted to the hospital. An- other limitation is that diagnosis and external cause of injury codes may change over time [34,35]. It is possible that systematic underreporting of football-related injuries is present in the NEDS, in part, because the football-specific E-code was first introduced in 2009 and its use may not have been universally adopted in the years studied. Furthermore, administrative databases have been shown to be subject to systematic underreporting for some factors [35]. In addition, the NEDS data are based on admission and patients who may have presented with repeat- ed injuries could not be identified.

Conclusion

In this nationally representative sample, most ED-treated injuries as- sociated with football were not acutely life threatening and very few re- quired major therapeutic intervention. Across all age groups, most injured patients were treated in the ED and discharged directly to home. This study provides a cross-sectional overview of ED presenta- tion for acute football-related injury across age groups at the population level in recent years. Longitudinal studies may be warranted to examine associations between the patterns of injury observed in this study and long-term outcomes among American tackle football players.

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ajem.2016.02.057.

References

  1. Bureau USC. Table 1249 participation in selected sports activities: 2009. Statistical Abstract of the United States; 2012. p. 768.
  2. Associations. NFoSHS. NFHS Participation Figures Search. Available at: http://www. nfhs.org/ParticipationStatics/ParticipationStatics.aspx/. [Accessed March 3, 2015].
  3. Delaney JS. Head injuries presenting to emergency departments in the United States from 1990 to 1999 for ice hockey, soccer, and football. Clin J Sport Med 2004;14(2): 80-7.
  4. Monroe KW, Thrash C, Sorrentino A, King WD. Most common sports-related injuries in a pediatric emergency department. Clin Pediatr (Phila) 2011;50(1):17-20.
  5. Powell JW, Barber-Foss KD. injury patterns in selected high school sports: a review of the 1995-1997 seasons. J Athl Train 1999;34(3):277-84.
  6. Shankar PR, Fields SK, Collins CL, Dick RW, Comstock RD. Epidemiology of high school and collegiate football injuries in the United States, 2005-2006. Am J Sports Med 2007;35(8):1295-303.
  7. Adickes MS, Stuart MJ. Youth football injuries. Sports Med 2004;34(3):201-7.
  8. Ramirez M, Schaffer KB, Shen H, Kashani S, Kraus JF. Injuries to high school football athletes in California. Am J Sports Med 2006;34(7):1147-58.
  9. Saal JA. Common American football injuries. Sports Med 1991;12(2):132-47.
  10. Fainaru S, Fainaru-Wade M. Youth football participation drops. ESPN 2013.
  11. Hendricks M. ‘Early’ NFL retirements show changing tide in the NFL. USA Today; 2015.
  12. Belson K. Chris Borland, fearing for health, retires from the 49ers. At 24. The New York Times; 2015.
  13. Stamm JM, Bourlas AP, Baugh CM, Fritts NG, Daneshvar DH, Martin BM, et al. Age of first exposure to football and later-life cognitive impairment in former NFL players. Neurology 2015;84(11):1114-20.
  14. Randolph C, Karantzoulis S, Guskiewicz K. Prevalence and characterization of mild cognitive impairment in retired National Football League players. J Int Neuropsychol Soc 2013;19(8):873-80.
  15. Guskiewicz KM, Marshall SW, Bailes J, McCrea M, Cantu RC, Randolph C, Jordan BD, et al. Association between recurrent concussion and late-life cognitive impairment in retired professional football players. Neurosurgery 2005;57(4):719-26 [discus- sion 719-726].
  16. Schwenk TL, Gorenflo DW, Dopp RR, Hipple E. Depression and pain in retired profes- sional football players. Med Sci Sports Exerc 2007;39(4):599-605.
  17. Strain J, Didehbani N, Cullum CM, Mansinghani S, Conover H, Kraut MA, et al. De- pressive symptoms and white matter dysfunction in retired NFL players with con- cussion history. Neurology 2013;81(1):25-32.
  18. Hart Jr J, Kraut MA, Womack KB, Strain J, Didehbani N, Bartz E, et al. Neuroimaging of cognitive dysfunction and depression in aging retired National Football League players: a cross-sectional study. JAMA Neurol 2013;70(3):326-35.
  19. Golightly YM, Marshall SW, Callahan LF, Guskiewicz K. Early-onset arthritis in re- tired National Football League players. J Phys Act Health 2009;6(5):638-43.
  20. Cottler LB, Ben Abdallah A, Cummings SM, Barr J, Banks R, Forchheimer R. Injury, pain, and prescription opioid use among former National Football League (NFL) players. Drug Alcohol Depend 2011;116(1-3):188-94.
  21. Badgeley MA, McIlvain NM, Yard EE, Fields SK, Comstock RD. Epidemiology of 10,000 high school football injuries: patterns of injury by position played. J Phys Act Health 2013;10(2):160-9.
  22. Nation AD, Nelson NG, Yard EE, Comstock RD, McKenzie LB. Football-related injuries among 6- to 17-year-olds treated in US emergency departments, 1990-2007. Clin Pediatr (Phila) 2011;50(3):200-7.
  23. Mello MJ, Myers R, Christian JB, Palmisciano L, Linakis JG. Injuries in youth football: national emergency department visits during 2001-2005 for young and adolescent players. Acad Emerg Med 2009;16(3):243-8.
  24. Dick R, Ferrara MS, Agel J, Courson R, Marshall SW, Hanley MJ, et al. Descriptive ep- idemiology of collegiate men’s football injuries: National Collegiate Athletic Associa- tion Injury Surveillance System, 1988-1989 through 2003-2004. J Athl Train 2007; 42(2):221-33 [PubMed PMID: 17710170; PubMed Central PMCID: PMC1941296].
  25. HCUP Nationwide Emergency Department Sample (NEDS). Healthcare Cost and Uti- lization Project (HCUP). 2007, 2009. Agency for Healthcare Research and Quality, Rockville, MD. [accessed: May 2016, last modified: January 4, 2016] Available at: http://www.hcup-us.ahrq.gov/nedsoverview.jsp.
  26. Irick E. NCAA sports sponsorship and participation rates report. Indianapolis, Indiana: The National Collegiate Athletic Association; 2014 46206-22.
  27. Taylor BL, Attia MW. Sports-related injuries in children. Acad Emerg Med 2000; 7(12):1376-82.
  28. Knowles SB, Marshall SW, Bowling MJ, Loomis D, Millikan R, Yang J, Mueller FO. Risk factors for injury among high school football players. Epidemiology 2009;20(2): 302-10.
  29. Guskiewicz KM, Weaver NL, Padua DA, Garrett Jr WE. Epidemiology of concussion in collegiate and high school football players. Am J Sports Med 2000;28(5):643-50.
  30. US Census Bureau. Intercensal Estimates. Available: http://www.census.gov/popest/ data/intercensal [accessed: May 2016, last modified: October 9, 2012].
  31. Gessel LM, Fields SK, Collins CL, Dick RW, Comstock RD. Concussions among United

    States high school and collegiate athletes. J Athl Train 2007;42(4):495-503 [PubMed PMID: 18174937; PubMed Central PMCID: PMC2140075].

    Kucera KL, Klossner D, Colgate B, Cantu RC. National Center for Catastrophic Sport Injury Research: Annual Survery of Football Injury Research, Report #:2015-01, Available: https://nccsir.unc.edu/files/2013/10/Annual-Football-2014-Fatalities- Final.pdf. [accessed: May 2016, Final: March 2015].

  32. Darrow CJ, Collins CL, Yard EE, Comstock RD. Epidemiology of severe injuries among United States high school athletes: 2005-2007. Am J Sports Med 2009;37(9): 1798-805. http://dx.doi.org/10.1177/0363546509333015 [Epub 2009 Jun 16.

    PubMed PMID: 19531659].

    Sarrazin MS, Rosenthal GE. Finding pure and simple truths with administrative data.

    JAMA 2012;307(13):1433-5.

    Haut ER, Pronovost PJ, Schneider EB. Limitations of administrative databases. JAMA 2012;307(24):2589 [author reply 2589-2590].