Population-based rate of presentation to the emergency department (ED) for primary diagnosis of sports-related ocular injury by sex and age, from January 1, 2010, to December 31, 2013.
Aggregate number of presentations to the emergency department (ED) for sports-related ocular injury by mechanism of injury and primary vs secondary diagnosis, from January 1, 2010, to December 31, 2013.
eTable. External Cause of Injury Codes (E-Codes) Used for Identification of Sports-Related Ocular Trauma
Customize your JAMA Network experience by selecting one or more topics from the list below.
Haring RS, Sheffield ID, Canner JK, Schneider EB. Epidemiology of Sports-Related Eye Injuries in the United States. JAMA Ophthalmol. 2016;134(12):1382–1390. doi:10.1001/jamaophthalmol.2016.4253
Copyright 2016 American Medical Association. All Rights Reserved.
What activities led to the highest number of sports-related eye injuries between 2010 and 2013 in the United States?
In a cross-sectional analysis of a nationally representative emergency department database, 120 847 individuals presented to emergency departments nationwide between 2010 and 2013 with a diagnosis of sports-related ocular injury, which was the primary diagnosis in 85 961 patients. Basketball, baseball, and air guns were the most common causes of injury, accounting for almost half of all primary sports-related eye injuries.
These findings may help set priorities and develop strategies to try to reduce sports-related eye injuries among individuals presenting to emergency departments in the United States each year.
Ocular trauma can lead to lifelong sequelae, and sports-related ocular injuries have been shown to disproportionately affect the young. Studies quantifying and characterizing the incidence and type of injuries seen with sports-related ocular trauma may be useful for resource utilization, training, and prevention efforts.
To examine the emergency department (ED) burden of sports-related ocular trauma in the United States.
Design, Setting, and Participants
This retrospective, cross-sectional study examined the Nationwide Emergency Department Sample, containing data from approximately 30 million ED visits annually at more than 900 hospitals nationwide, from January 1, 2010, to December 31, 2013, to determine factors associated with sports-related ocular trauma.
Main Outcomes and Measures
Annual incidence of sports-related ocular trauma, broken down by age, sex, mechanism of injury, and related activity, as well as factors associated with short-term impaired vision.
During the study period, 120 847 individuals (mean age, 22.3 years [95% CI, 21.9-22.7]; 96 872 males, 23 963 females, and 12 with missing data) presented with sports-related ocular trauma, which was the primary diagnosis in 85 961 patients. Injuries occurred most commonly among males (69 849 [81.3%]; 95% CI, 80.6%-81.9%) and occurred most frequently as a result of playing basketball (22.6%; 95% CI, 21.7%-23.6%), playing baseball or softball (14.3%; 95% CI, 13.7%-14.9%), and shooting an air gun (11.8%; 95% CI, 10.8%-12.8%). Odds of presentation to the ED with impaired vision were greatest for paintball and air gun injuries relative to football-related injuries (odds ratio, 4.75; 95% CI, 2.21-10.19 and 3.71; 95% CI, 2.34-5.88, respectively; P < .001).
Conclusions and Relevance
In our study, approximately 30 000 individuals presented annually to EDs in the United States with sports-related eye injuries; in more than 70% of these cases, eye injuries were the primary diagnosis. Activities involving projectiles pose the greatest risk for visual impairment in the short term, although long-term outcomes were unavailable.
Ocular injuries are a significant cause of morbidity and disability in the US population.1,2 The Centers for Disease Control and Prevention estimates that the annual rate of presentation to the emergency department (ED) for all-cause ocular injury is approximately 37.6 per 10 000 population.3 Eye injuries can have long-term sequelae that affect quality of life for years and can predispose the individual to further injury, depression, and systemic disease.4-15 Visual impairment also poses substantial economic challenges to patients, health systems, and payers, which may further affect the patient’s access to quality care and follow-up.16-18
Sports-related activities contribute to a substantial proportion of injuries in the ED, particularly among young people.19-22 Although many of these injuries are musculoskeletal in nature, sports-related injuries to the central nervous system and eyes are not uncommon in EDs.21,23-30 Eye injuries sustained in sports-related activities may differentially affect young people, predisposing them to the possibility of long-term sequelae and reduced quality of life.20,29,31-33
Previous efforts to estimate the national burden of sports-related eye injuries have been hampered by inconsistencies in how these injuries have been defined; for example, narrow definitions of the word sport, possibly influenced by regional and cultural differences, have exaggerated interstudy variability.5,34-37 Unclear definitions of delineations between organized sporting activities and recreational sporting activities contribute to this variability. A 2013 article by Selassie et al38 expanded the accepted definition of sports-related injuries; using external cause of injury codes from the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), the authors were able to include injuries sustained during recreational activities, such as swimming, running, and off-road power and equestrian sports. Modifications to the ICD-9-CM implemented in 2009 further expanded researchers’ ability to characterize sports-related injury by differentiating between specific team sports (eg, football, baseball, or basketball).39
In light of these developments, we sought to estimate and characterize the burden of sports-related ocular trauma in EDs in the United States from 2010 to 2013.
The Nationwide Emergency Department Sample (NEDS), part of the Healthcare Cost and Utilization Project, is the nation’s largest all-payer ED database; as such, it includes data on all patients, regardless of insurance status. When weighted to the level of the US population, the NEDS contains administrative discharge data on approximately 30 million ED visits annually from more than 900 hospitals nationwide.40 The NEDS and other Healthcare Cost and Utilization Project data sets are often used by the National Center for Injury Prevention and Control for the purpose of epidemiologic statistics and surveillance, and compare favorably with other well-established national samples, such as the National Hospital Ambulatory Medical Care Survey and the National Hospital Discharge Survey.41,42
Ocular trauma was defined as a diagnosis of either blowout fracture of the orbit, open wounds of the ocular adnexa or eye, superficial injury or contusion to the eye or adnexa, foreign body on the external eye, burn confined to the eye and adnexa, or injury to the optic nerve and pathways or cranial nerves III, IV, or VI (ICD-9-CM codes 802.6, 802.7, 870.x, 871.x, 918.x, 921.x, 930.x, 940.x, 950.x, 951.0, 951.1, and 951.3). These cases were further identified as either a primary or nonprimary diagnosis of sports-related ocular trauma. Although the acute and long-term effects of ocular injuries can be just as severe or worse for patients for whom the eye-related injury is not the primary diagnosis, including those associated with motor vehicle crashes, traumatic brain injury, and other polytraumatic injuries, we focused much of our analysis on patients with primary diagnoses of ocular injury to more clearly elucidate which activities contribute most specifically to eye injury. Patients with sports-related ED presentation were identified using a list of ICD-9-CM external cause of injury codes compiled from examination of the literature and ICD-9-CM codes available for the dates of interest (eTable in the Supplement). This study was approved by the Johns Hopkins Medical Institutions institutional review board. The NEDS is an administrative limited data set and individual patient consent was not required.
We queried the NEDS database using the above criteria to identify all patients presenting to EDs in the United States between January 1, 2010, and December 31, 2013; before 2010, details on injuries sustained while playing specific team sports (eg, football, baseball, basketball) were either unavailable or not widely used, which limited their utility for the purposes of this study. Data examined included visit-level data, such as information related to the injury, diagnosis, ED-specific charges, geographic region, and expected primary insurer, as well as patient-level data, such as age, sex, and zip code–based estimated income quartile. χ2 Testing was performed on these categorical variables to measure association, with P ≤ .05 considered significant. Population-specific rates were calculated using data from the US Census Bureau.
The NEDS does not contain data on visual acuity, and although diagnoses of visual abnormalities may represent preexisting conditions, the likelihood that participants in sports would have such a preexisting diagnosis is low. Multivariable logistic regression modeling was used to explore the effect of patient- and injury-level factors on the odds of impaired vision, which included visual disturbances (ICD-9-CM code 368.x) and blindness and low vision (ICD-9-CM code 369.x); as with others, these diagnoses are only present in the medical record if they were made at the time of the ED visit, so lack of a vision-related code may reflect either unimpaired vision or that assessment of vision was not performed. As this model was explanatory in nature, stepwise variable selection for parsimony was not used. Wald tests were used to assess the effect of individual variables on the model, and P ≤ .05 was considered significant.
Standard descriptive statistical methods were used to analyze associations between patient, injury, and hospital characteristics and presentation to the ED with sports-related ocular injuries.
During the 4-year study period, 120 847 patients presented to the ED with sports-related ocular trauma; for 85 961 of these individuals, ocular trauma was the primary diagnosis. These 85 961 patients with primary sports-related ocular trauma accounted for approximately 3.3% of all 2 636 037 patients with primary ocular trauma presenting to the ED during the study period, and the 60 027 patients with isolated sports-related ocular trauma represented 2.8% of all 2 156 962 patients with isolated ocular trauma (no other injuries reported). Patients with primary sports-related ocular trauma were most commonly male (69 849 [81.3%]; 95% CI, 80.6%-81.9%). Mean patient ages were 20.1 years (95% CI, 19.7-20.5) and 19 years (95% CI, 18.4-19.7) for males and females, respectively (P < .001). Age-specific rates of injury increased for every successive year from age 7 until age 15 years (among females) or age 17 years (among males), after which they were markedly reduced (Figure 1). More than half of the males (41 775 [59.8%]) and females (10 814 [67.1%]) who sustained sports-related primary ocular injuries were 18 years or younger.
Basketball was the leading cause of injury among male patients (25.7%; 95% CI, 24.7%-26.7%), followed by baseball or softball (13.2%; 95% CI, 12.5%-13.8%) and shooting an air gun (12.7%; 95% CI, 11.7%-13.8%) (Table 1). Baseball or softball was the most common cause of injury among female patients (19.2%; 95% CI, 17.8%-20.7%), followed by cycling (10.8%; 95% CI, 9.7%-12.1%) and soccer (10.3%; 95% CI, 9.2%-11.5%). Although cycling accounted for the second-highest total number of eye injuries (n = 21 227), less than half of these injuries (8319 [39.2%]) included a primary diagnosis of ocular trauma; the same was true for ocular injuries associated with equestrian sports (310 [46.4%]) and off-road power sports (1873 [35.4%]). The sport-specific proportion of ocular injuries identified as primary vs secondary varied by sport, with soccer (5688 [92.3%]), shooting an air gun (10 101 [91.1%]), and shooting a paintball gun (763 [89.7%]) demonstrating the largest proportions of patients presenting with sport-specific ocular injuries identified as primary (Figure 2).
Overall, open wounds of the adnexa (ICD-9-CM code 870.x) were the most common type of injury (33.5%; 95% CI, 32.4%-34.5%), followed by contusions of the eye and adnexa (ICD-9-CM code 921.x; 30.1%; 95% CI, 29.1%-31.1%) and superficial injuries of the eye and adnexa (ICD-9-CM code 918.x; 21.1%; 95% CI, 20.4%-21.8%) (Table 1). Among basketball-related injuries, open wounds of the adnexa were most common (9062 [46.6%]; 95% CI, 44.7%-48.4%), followed by superficial wounds (5672 [29.1%]; 95% CI, 27.6%-30.7%); for baseball-related incidents, contusions (6331 [51.5%]; 95% CI, 49.4%-53.7%) were most common, followed by orbital fractures (2568 [20.9%]; 95% CI, 19.2%-22.7%). Injuries sustained from shooting an air gun were most commonly contusions (4818 [47.7%]; 95% CI, 44.9%-50.5%), followed by superficial wounds of the eye and adnexa (2946 [29.2%]; 95% CI, 26.9%-31.5%).
Although only 3760 (3.1%) of all individuals who presented to the ED with a diagnosis of ocular trauma had impaired vision, rates of visual symptoms varied widely by sports-related activity, injury type, and age. The proportion of injuries resulting in impaired vision was highest for those injuries associated with paintball (10.2%), shooting an air gun (8.2%), racket sports (5.8%), and soccer (5.7%) (Table 2). After adjustment for other variables, the odds of impaired vision for sports-related ocular trauma were 4.75 times higher for a paintball-related injury (95% CI, 2.21-10.19) and 3.71 times higher for an air gun–related injury (95% CI, 2.34-5.88) than for football-related ocular injuries (P < .001) (Table 3). Adjusted odds of visual impairment generally increased by age, and did not vary by sex, income quartile, or season.
Traumatic ocular events and their sequelae predispose patients to long-term physical, psychosocial, and economic effects, and sports-related events have been reported as a prominent contributor to the overall burden of ocular trauma.9,10,12,43 Sports-related ocular trauma has been identified as a substantial risk factor for inpatient hospitalization, with some estimates suggesting that 22% to 40% of hospitalizations for ocular trauma were the result of injuries sustained during participation in sporting activities.25,26,44
We have found that these injuries represent a substantial burden in EDs in the United States, accounting for approximately 30 000 ED visits annually—an estimate substantially higher than previously reported.5 Presenting patients tended to be young, and incidence peaked during adolescent years for both male and female patients. This differential burden on the young highlights the potential for long-term loss of quality-adjusted life years.
Patients presenting with injuries resulting from basketball, cycling, and baseball or softball and air guns represented the largest cohort of ED presentations for all primary and nonprimary sports-related ocular injuries. Among only patients with primary ocular injuries, ED presentation was most commonly owing to basketball, baseball or softball, and shooting an air gun. Combined, these 3 activities accounted for 48.7% of all primary injuries in this population. Although most injuries resulting from sports-related activities were superficial, more than one-fifth of baseball-related injuries were blowout fractures of the orbit.
Impaired vision was rare but showed a strong affiliation with recreational projectile-firing devices. Paintball and air guns accounted for 991 of all 3760 cases resulting in impaired vision (26.4%), despite accounting for only 11 937 of all 120 847 injuries (9.9%). Soccer-related injuries also contributed a disproportionate share of visual impairment (408 of all 3760 vision impairments [10.9%] and 7162 of all 120 847 injuries [5.9%]). Cycling-related injuries were relatively less likely to lead to visual impairment (377 of all 3760 visual impairments [9.7%] and 21 227 of all 120 847 injuries [17.6%]).
Logistic regression modeling underscored the elevated risk associated with projectile-firing recreational devices, as injuries from shooting paintball and air guns showed 4.75 and 3.71 times greater adjusted odds, respectively, of leading to visual impairment than injuries sustained during participation in football. Patients with injuries from soccer and hockey also showed substantially elevated risk of vision impairment after adjustment, compared with patients presenting to the ED with isolated ocular injuries sustained playing football (odds ratio, 2.42 and 2.32, respectively; P < .001) (Table 3).
Use of an administrative database such as the NEDS carries several limitations.45 The number of individuals sustaining sports-related ocular injuries who present to non-ED sources, such as primary care physicians, outpatient ophthalmologists, and urgent care facilities, or who choose not to seek care, is unknown, and such individuals will not be included in ED-specific data sets such as the NEDS; our study is limited to those presenting to the ED for care. Similarly, data after discharge from the ED, such as surgical intervention and length of hospitalization, are unavailable in this data set, as are data on long-term visual outcomes or detailed assessments of visual acuity, all of which would likely be of interest to both health care professionals and policy makers. The NEDS contains administrative discharge data as provided by participating hospitals to the Healthcare Cost and Utilization Project via state agencies; while coding practices at individual hospitals may vary to some extent, it is anticipated that the treating health care professional would generally be responsible for diagnoses.
These results suggest a path forward for efforts to prevent sports-related ocular trauma. Research has repeatedly shown that appropriate protective eyewear can reduce the incidence of sports-related ocular trauma.37,46-51 Mandating the use of such protective gear has reduced rates of injury across several sports, and recent research suggests that when appropriate eyewear is available but not mandatory, top-performing athletes frequently choose to wear it.48
Although such approaches may be appropriate for organized team sports, they are not likely to affect the burden of sports-related ocular trauma due to nonorganized recreational sporting activities, including those most correlated with visual impairment, such as shooting paintball and air guns. Reducing sports-related ocular trauma among individuals engaging in these activities, along with individual sports with high levels of injuries, such as cycling, will likely require a coordinated approach from policy makers, industry, and public health professionals. Further research is warranted to identify opportunities for intervention among sports participants.
Accepted for Publication: September 7, 2016.
Corresponding Author: R. Sterling Haring, DO, MPH, Center for Healthcare Quality and Patient Safety, University of Lugano, Via G. Buffi 13, 6900 Lugano, Switzerland (firstname.lastname@example.org).
Published Online: November 3, 2016. doi:10.1001/jamaophthalmol.2016.4253
Author Contributions: Dr Haring had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Haring, Sheffield, Schneider.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Haring.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Haring, Sheffield, Schneider.
Administrative, technical, or material support: Haring, Canner, Schneider.
Study supervision: Schneider.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Previous Presentations: These findings were presented at the Academic Surgical Congress of the Association for Academic Surgery; February 2, 2016; Jacksonville, Florida.
Create a personal account or sign in to: