Emergency Department Visits With Facial Fractures Among Children and Adolescents: An Analysis of Profile and Predictors of Causes of Injuries

Transcription

1 CRANIOMAXILLOFACIAL TRAUMA Emergency Department Visits With Facial Fractures Among Children and Adolescents: An Analysis of Profile and Predictors of Causes of Injuries Veerasathpurush Allareddy, BDS, MBA, MHA, MMSc, PhD,* Abraham Itty, DMD,y Elyse Maiorini, DMD,z Min Kyeong Lee, DMD,x Sankeerth Rampa, MBA, MPH,k Veerajalandhar Allareddy, MD, MBA,{ and Romesh P. Nalliah, BDS# Purpose: The objectives of this study were to provide nationally representative estimates of hospitalbased emergency department (ED) visits for facial fractures in children and adolescents, examine the burden associated with such visits, identify common types of facial fracture, and examine the role of patient-related demographic factors on the causes of facial fractures. Materials and Methods: The Nationwide Emergency Department Sample for 2008 to 2010 was used. All ED visits with a diagnosis of facial fractures in those no older than 21 years were selected. Demographic characteristics, types of facial fracture, causes of injuries, and hospital charges were examined. Results: During the study period, 336,124 ED visits were for facial fractures in those no older than 21 years. Late adolescents (18 to 21 yr old) and middle adolescents (15 to 17 yr old) comprised 45.6% and 26.6% of all ED visits, respectively. Male patients comprised 74.7% of ED visits. The most common facial fractures were those of the nasal bones and mandible. Younger children were more likely to have falls, pedal cycle accidents, pedestrian accidents, and transport accidents, whereas older groups were more likely to have firearm injuries, motor vehicle traffic accidents, and assaults (P <.05). Female patients were more likely to have falls, motor vehicle traffic accidents, and transport accidents, whereas male patients were more likely to have firearm injuries, pedal cycle accidents, and assaults (P <.05). Those residing at low annual income household levels were at a high risk for having firearm injuries, motor vehicle traffic accidents, and transport accidents (P <.05). Conclusions: Late adolescents, middle adolescents, and male patients comprise a significant proportion of these ED visits. Age, gender, and household income levels are significantly associated with the causes of facial fracture injuries. Ó 2014 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 72: , 2014 *Associate Professor, Department of Orthodontics, College of Dentistry University of Iowa, Iowa City, IA. ypediatric Dental Resident, Texas A&M University Baylor College of Dentistry, Dallas, TX. zgeneral Practice Resident, Carolinas Medical Center, Charlotte, NC. xorthodontic Resident, Department of Developmental Biology, Harvard School of Dental Medicine, Boston, MA. kresearch Assistant, Department of Pediatric Critical Care, Case Medical School Rainbow Babies Hospital, Cleveland, OH. {Assistant Professor, Department of Pediatric Critical Care, Case Medical School Rainbow Babies Hospital, Cleveland, OH. Unintentional injuries are the leading causes of mortality in those 1 to 24 years old in the United States. 1 Each year, a substantial number of children seek hospitalbased emergency departments (EDs) for treatment of traumatic nonfatal injuries. 2 Estimates suggest that only a small proportion of these injuries involve the #Instructor, Office of Global and Community Health, Harvard School of Dental Medicine, Boston, MA. Conflict of Interest Disclosures: None of the authors reported any disclosures. Address correspondence and reprint requests to Dr Nalliah: Office of Global and Community Health, Harvard School of Dental Medicine, Boston, MA 02115; romesh.aus@gmail.com Received January Accepted March Ó 2014 American Association of Oral and Maxillofacial Surgeons /14/ $36.00/

2 ALLAREDDY ET AL 1757 maxillofacial region. 3 Reports from the Nationwide Inpatient Sample and Kids Inpatient Dataset suggest that the pediatric population accounts for approximately 15% of all facial fractures. 4 It has been hypothesized that children possess certain anatomic attributes that make them less prone to facial fractures compared with adults. 5 However, when children do sustain facial fractures, they may have several concomitant injuries that require visits to hospital-based EDs and subsequent hospitalization. Previous published studies in the literature focusing on pediatric facial fractures have provided estimates of patient populations that are restricted to single or regional hospital settings. 6,7 Very few studies have used national registries. 4,8 There is a paucity of nationally representative generalizable estimates of profile and high-risk cohorts among children and younger cohorts who are likely to have facial fractures and present to hospital-based EDs. The objectives of this descriptive epidemiologic study were to provide nationally representative estimates of hospital-based ED visits for facial fractures in children and adolescents (#21 yr old), examine the financial burden associated with such visits, identify common types of facial fracture, and examine the role of patient-related demographic factors on the causes of facial fractures. Materials and Methods DESIGN AND DATABASE DESCRIPTION The authors performed a retrospective analysis of the Nationwide Emergency Department Sample (NEDS) for 2008 to The NEDS is part of a family of databases developed for the Healthcare Cost and Utilization Project (HCUP) sponsored by the Agency for Healthcare Research and Quality (AHRQ). 9 The NEDS is the largest all-payer ED database in the United States, yielding national weighted estimates of roughly 130 million hospital-based ED visits. Each ED visit in the NEDS database has an assigned sample weight and this was used to project all estimates to national levels. VARIABLES IN NEDS The NEDS is composed of more than 100 clinical and nonclinical variables for each hospital stay, including demographic data such as hospital and patient characteristics, geographic area, and the nature of the ED visits (eg, common reasons for ED visits, including injuries). Identification of injuryrelated ED visits, including mechanism, intent, and severity of injury, is available. Patient demographic characteristics (eg, gender, age, urban vs rural designation of residence, national quartile of median household income for patient s zip code) and hospital characteristics (eg, region, urban vs rural location, teaching status) are available. PRIVACY, DATA USER AGREEMENT, AND INSTITUTIONAL REVIEW BOARD The NEDS has safeguards to protect the privacy of individual patients, physicians, and hospitals. In accord with university hospitals, the Case Medical Center, and institutional review board and in agreement with Federal Regulation 45 CFR (b) ( research involving the collection or study of existing data, documents, records, pathological specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects ), such studies are permitted to be classified as research that is exempt from full or expedited review by the institutional review board. One of the authors (V.J.A.) completed a data user agreement with the HCUP and AHRQ and obtained the datasets. According to the data user agreement, cell counts of 10 and lower cannot be reported to maintain patient privacy. Hence, low cell counts are not reported. Instead, the term DS (discharge information suppressed) has been used to indicate low cell counts. CASE SELECTION All ED visits concerning children and adolescents (#21 yr old) with diagnostic codes for facial fractures were selected for analysis. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) was used to identify different types of fracture and included closed fracture of nasal bones (802.0), open fracture of nasal bones (802.1), closed mandibular fracture combined, open mandibular fracture combined, closed fracture of an unspecified site of the mandible (802.20), closed fracture of the condylar process of the mandible (802.21), closed fracture of the subcondylar process of the mandible (802.22), closed fracture of the coronoid process of the mandible (802.23), closed fracture of an unspecified part of the ramus of the mandible (802.24), closed fracture of the angle of the jaw (802.25), closed fracture of the symphysis of the body of the mandible (802.26), closed fracture of the alveolar border of the body of the mandible (802.27), closed fracture of other and unspecified part of the body of the mandible (802.28), closed fracture of multiple sites of the mandible (802.29), open fracture of an unspecified site of the mandible (802.30), open fracture of the condylar process of the mandible (802.31), open fracture of the subcondylar process of the mandible (802.32), open fracture of the coronoid process of the mandible (802.33), open fracture of an unspecified

3 1758 PEDIATRIC EMERGENCY VISITS FOR FACIAL FRACTURE part of the ramus of the mandible (802.34), open fracture of the angle of the jaw (802.35), open fracture of the symphysis of the body of the mandible (802.36), open fracture of the alveolar border of the body of the mandible (802.37), open fracture of the body of other and unspecified part of the mandible (802.38), open fracture of multiple sites of the mandible (802.39), closed fracture of the malar and maxillary bones (802.4), open fracture of the malar and maxillary bones (802.5), closed fracture of the orbital floor (blowout; 802.6), open fracture of the orbital floor (blowout; 802.7), closed fracture of other facial bones (802.8), and open fracture of other facial bones (802.9). External cause-of-injury codes (E-Codes) were used to identify the causes of facial fractures. OUTCOME VARIABLES EXAMINED Characteristics of all ED visits and subsequent hospitalizations, including age, gender, insurance status, causes of facial fractures, disposition of the patient from the ED, and disposition of the patient after inpatient admission into the same hospital, were examined. Age was divided into 7 groups infants (<1 yr old), toddlers (1 to 3 yr old), preschool (4 to 5 yr old), school-age (6 to 10 yr old), early adolescents (11 to 14 yr old), middle adolescents (15 to 17 yr old), and late adolescents (18 to 21 yr old) based on the Bright Futures: Guidelines for Health Supervision of Infants, Children, and Adolescents recommended and published by the American Academy of Pediatrics. 10 ED charges, hospitalization charges, and length of stay also were examined. All charges were adjusted for inflation to 2010 levels using the Bureau of Labor Statistics inflation rates for in-hospital care. 11 ANALYTICAL APPROACH Descriptive statistics were used to summarize the data. SAS procedure codes proc surveyfreq and proc surveymeans were used for descriptive analysis. For all analyses, the NEDS hospital stratum was the stratification unit and each ED visit was the unit of analysis. Different causes of injuries (falls, firearms accidents, motor vehicle traffic accidents, pedal cycle accidents, pedestrian accidents, transport accidents, and assaults) were modeled as outcome variables. The influence of patient-related factors, including age, gender, and household income levels (based on zip code), on different causes of injuries was examined. Because causes of injuries were coded as binomial variables, multivariable logistic regression analyses were used to fit the models. The SAS procedure code proc rlogist was used for multivariable logistic regression analyses. A separate regression model was used for each cause. The default setting in SAS callable SUDAAN (Taylor linearization methods) was used to compute the variances. A with replacement method was used in the multivariable regression models. Odds ratios and associated 95% confidence intervals were computed for each level of patient factor. It is likely that patients treated in a hospital are subject to similar processes of care, which would result in clustering of outcomes within hospitals. The authors adjusted for this effect (within-hospital effect) by using a nest statement. The stratum to which the hospital belonged and the hospital identifier were used in the nest statement to account for the clustering effects within hospitals. There is a possibility of clustering of outcomes within cities. Because the current dataset does not include all hospitals in a particular city, the authors were unable to adjust for this effect. All statistical analyses were 2-sided and a P value less than.05 was deemed statistically significant. All statistical analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC) and SAS callable SUDAAN (RTI International, Research Triangle Park, NC). Results During the study period, 336,124 ED visits were for facial fractures in those no older than 21 years. The characteristics of ED visits are listed in Table 1. The mean age at the ED visits was close to 16 years. Late adolescents (18 to 21 yr old) and middle adolescents (15 to 17 yr old) comprised 45.6% and 26.6% of all ED visits, respectively. Male patients comprised 74.7% of ED visits. The predominating primary payer was private insurance (51.6% of all ED visits). Approximately 18.2% of ED visits were uninsured. The 5 most frequently reported facial fractures were closed fracture of the nasal bones (59.8%), closed mandibular fracture (15.3%), closed fracture of other facial bones (13.3%), closed fracture of the orbital floor (blowout; 10.1%), and closed fracture of the malar and maxillary bones (8.3%). The frequently reported causes of injuries (Table 2) included injury from assaults (52.8%), falls (11.4%), motor vehicle traffic accidents (10.9%), pedal cycle accidents (2.7%), transport accidents (2.4%), firearms (0.6%), and pedestrian accidents (0.08%). Results of multivariable logistic regression analyses examining the association between patient characteristics (age, gender, and household income levels) and causes of injuries are presented in Table 3. Compared with those 18 to 21 years old, those in younger groups were more likely to have higher odds for falls, pedal cycle accidents, pedestrian accidents, and transport accidents (P <.001). The younger groups were more likely to have lower odds for firearm injuries, motor vehicle traffic accidents, and assaults (P <.001) compared with those 18 to 21 years old. Female patients were associated with higher odds for falls,

4 ALLAREDDY ET AL 1759 Table 1. CHARACTERISTICS OF HOSPITAL-BASED ED VISITS WITH FACIAL FRACTURES (N = 336,124) Characteristic ED Visits, % Age (yr) Mean (SEM) 15.9 (0.1) Infants (#1 yr) 1.0 Toddlers (>1-3 yr) 1.9 Preschool (4-5 yr) 2.3 School age (6-10 yr) 8.1 Early adolescents (11-14 yr) 14.6 Middle adolescents (15-17 yr) 26.6 Late adolescents (18-21 yr) 45.6 Gender Male 74.7 Female 25.3 Insurance Medicare 0.4 Medicaid 23.9 Private insurance 51.6 Uninsured 18.2 Other insurance plans 5.9 Types of facial fractures (ICD-9-CM codes) Closed fracture of nasal bones (802.0) 59.8 Open fracture of nasal bones (802.1) 2.0 Closed mandibular fracture, combined 15.3 Open mandibular fracture, combined 2.2 Closed fracture of unspecified site of 6.2 mandible (802.20) Closed fracture of condylar process of 1.7 mandible (802.21) Closed fracture of subcondylar process 1.0 of mandible (802.22) Closed fracture of coronoid process of 0.09 mandible (802.23) Closed fracture of unspecified part of 1.3 ramus of mandible (802.24) Closed fracture of angle of jaw (802.25) 2.7 Closed fracture of symphysis of body of 1.5 mandible (802.26) Closed fracture of alveolar border of 0.6 body of mandible (802.27) Closed fracture of other and 1.4 unspecified part of body of mandible (802.28) Closed fracture of multiple sites of 1.0 mandible (802.29) Open fracture of unspecified site of 0.6 mandible (802.30) Open fracture of condylar process of 0.09 mandible (802.31) Open fracture of subcondylar process 0.07 of mandible (802.32) Open fracture of coronoid process of 0.01 mandible (802.33) Open fracture of unspecified part of 0.2 ramus of mandible (802.34) Open fracture of angle of jaw (802.35) 0.5 Table 1. Cont d Characteristic ED Visits, % Open fracture of symphysis of body of 0.5 mandible (802.36) Open fracture of alveolar border of 0.1 body of mandible (802.37) Open fracture of body of other and 0.4 unspecified part of mandible (802.38) Open fracture of multiple sites of 0.2 mandible (802.39) Closed fracture of malar and maxillary 8.3 bones (802.4) Open fracture of malar and maxillary 0.4 bones (802.5) Closed fracture of orbital floor 10.1 (blowout) (802.6) Open fracture of orbital floor 0.2 (blowout) (802.7) Closed fracture of other facial bones 13.3 (802.8) Open fracture of other facial bones 0.5 (802.9) Income quartile (based on zip-code annual income household levels)* Quartile Quartile Quartile Quartile Year Abbreviations: ED, emergency department; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; SEM, standard error of the mean. * Income quartiles differ by year. For 2008, the levels were $1 to $38,999 (quartile 1), $39,000 to $48,999 (quartile 2), $49,000 to $63,999 (quartile 3), and at least $64,000 (quartile 4). For 2009, the levels were $1 to $39,999 (quartile 1), $40,000 to $49,999 (quartile 2), $50,000 to $65,999 (quartile 3), and at least $66,000 (quartile 4). For 2010, the levels were $1 to $40,999 (quartile 1), $41,000 to $50,999 (quartile 2), $51,000 to $66,999 (quartile 3), and at least $67,000 (quartile 4). Allareddy et al. Pediatric Emergency Visits for Facial Fracture. J Oral Maxillofac Surg motor vehicle traffic accidents, and transport accidents compared with male patients (P <.001). Those residing in lower annual household income quartiles were associated with lower odds for falls and assaults compared with those in the highest income level quartile (P <.001). Those residing in lower annual household income quartiles were associated with higher odds for firearm injuries, motor vehicle traffic accidents, and transport accidents (P <.05) compared with those in the highest income level quartile.

5 1760 PEDIATRIC EMERGENCY VISITS FOR FACIAL FRACTURE Table 2. CAUSES OF FACIAL FRACTURE AMONG ED VISITS ED Visits, % Characteristics All Patients #14 yr old (n = 93,956) Patients yr old (n = 89,380) Patients yr old (n = 153,148) External cause of injury (E-Codes) Fall Firearm injury Motor vehicle traffic accident Pedestrian accident Pedal cycle accident Transport accident Assault Abbreviation: ED, emergency department. Allareddy et al. Pediatric Emergency Visits for Facial Fracture. J Oral Maxillofac Surg Outcomes after the ED visit are presented in Table 4. Close to 81.2% were discharged routinely and 13.3% were admitted as inpatients to the same hospital as the ED visit. In total, 204 patients died in the ED. The mean ED charge per visit was $3,318 and the total ED charges across the entire United States was approximately $929 million. In total, 44,627 ED visits necessitated admission as inpatients into the same hospital (Table 5). The mean age was 16 years. Late adolescents and middle adolescents comprised 52.7% and 23.3% of hospitalizations, respectively. Most hospitalizations concerned male patients (76.1%). Private insurance was the predominating payer (48.6% of hospitalizations). Uninsured patients comprised 14.9% of hospitalizations. Commonly reported types of facial fracture in the hospitalized cohort included closed mandibular fractures (31.5%), closed fractures of the nasal bones (28.8%), closed fracture of other facial bones (24.2%), and closed fracture of the malar and maxillary bones (21.5%). The external causes of injuries resulting in hospitalizations included motor vehicle traffic accidents (36.6%), assaults (25.6%), falls (9.5%), transport accidents (5.8%), pedal cycle accidents (3.3%), firearms (3.1%), and pedestrian accidents (0.3%). After hospitalization, close to 84.5% were discharged routinely (Table 6). In total, 883 patients died in hospitals after admission as an inpatient. The total hospitalization charges and hospitalization days across the entire United States were close to $3.07 billion and 222,740 days, respectively. Discussion This study reviewed characteristics of ED visits and subsequent hospitalizations for facial fractures in children and adolescents no older than 21 years using a nationally representative hospital-based ED database. The study focused on 4 measurements: types of facial fracture, causes of pediatric facial fractures, disposition status after ED visits and subsequent hospitalization, and hospital charges (ED charges and hospitalization charges). In EDs, including trauma centers, oral and maxillofacial surgeons are consulted for facial injuries in all age groups. The present study showed that facial fractures are not an infrequent cause of ED visits and hospitalization in children and adolescents. This extensive descriptive review showed that most such injuries are minor and certain age groups are at increased risk of specific causes of injury. Although overall mortality is low, mortality in hospitalized children is not insignificant. The incidence of fractures of facial bones increases with increasing age and peaks in late adolescence (18 to 21 yr old). Although facial bone fractures are uncommon in children younger than 5 years, certain age groups appear to be at a higher risk of specific causes of such injuries. Facial fractures are frequent causes of ED visits and are associated with significant hospital resource usage. Tailoring preventive programs toward age group and cause of injury may decrease the incidence of such injuries. The present study estimates are consistent with those of previous studies for fracture locations, etiology of injuries, and age and gender distributions. The present study findings indicate that nasal fractures are the most common facial fracture in children and adolescents, comprising close to 60% of ED visits. The second most frequently occurring fracture was that of the mandible. These findings are consistent with those of previous reports obtained from the National Trauma Registry. 8 It should be noted that nasal bones are less resistant to fractures compared

6 Table 3. RESULTS OF MULTIVARIABLE LOGISTIC REGRESSION ANALYSIS EXAMINING ROLE OF PATIENT-RELATED FACTORS ON CAUSES OF INJURIES Characteristic Response Fall Firearm Injury Motor Vehicle Traffic Accident Causes of Injuries, OR (95% CI) Pedal Cycle Accident Pedestrian Accident Transport Accident Age # ( ) y 0.18 ( ) 0.49 ( ) y 0.83 ( ) 3.89 ( ) 0.19 ( )* 0.21 ( ) y > ( ) y 0.27 ( )* 0.54 ( ) y 1.46 ( ) 6.31 ( ) y 0.95 ( ) 0.31 ( ) y ( ) y 0.15 ( )* 0.68 ( ) y 5.81 ( ) y 7.17 ( ) y 1.64 ( ) y 0.44 ( ) y ( ) y 0.11 ( ) y 0.64 ( ) y 7.24 ( ) y 8.25 ( ) y 2.17 ( ) y 0.89 ( ) y ( ) y 0.25 ( ) y 0.45 ( ) y 4.46 ( ) y 2.08 ( ) 1.83 ( ) y 1.61 ( ) y ( ) y 0.54 ( ) y 0.77 ( ) y 1.58 ( ) y 1.38 ( ) 1.18 ( )* 1.45 ( ) y reference reference reference reference reference reference reference Gender Female 1.34 ( ) y 0.49 ( ) y 2.04 ( ) y 0.64 ( ) y 1.16 ( ) 1.32 ( ) y 0.69 ( ) y Male reference reference reference reference reference reference reference Income quartiles Quartile ( ) y 5.79 ( ) y 1.19 ( )* 0.77 ( ) y 1.17 ( ) 1.27 ( )* 0.84 ( ) y Quartile ( ) y 3.46 ( ) y 1.22 ( ) y 1.02 ( ) 1.05 ( ) 1.40 ( ) y 0.85 ( ) y Quartile ( ) y 2.31 ( ) y 1.21 ( ) y 1.03 ( ) 0.73 ( ) 1.32 ( ) y 0.88 ( ) y Quartile 4 reference reference reference reference reference reference reference Abbreviations: CI, confidence interval; OR, odds ratio. * P <.05. y P <.001. Allareddy et al. Pediatric Emergency Visits for Facial Fracture. J Oral Maxillofac Surg Assault ALLAREDDY ET AL 1761

7 1762 PEDIATRIC EMERGENCY VISITS FOR FACIAL FRACTURE Table 4. OUTCOMES AFTER ED VISITS ED Visits, % Outcomes All Patients #14 yr old (n = 93,956) Patients yr old (n = 89,380) Patients yr old (n = 153,148) Disposition from ED Routine Transfer to short-term hospital Transfer other (long-term care facilities, eg, skilled nursing facilities) Home health care Against medical advice Admitted as inpatient to this hospital Died in ED Not admitted to this hospital, destination unknown/others Total ED charge ($)* Mean (SEM) 3,318 (70.93) 2,689 (64.24) 3,122 (71.21) 3,819 (93.70) Total across entire United States 929,095, ,438, ,759, ,897,720 Abbreviations: ED, emergency department; SEM, standard error of the mean. * All charges were adjusted for inflation to 2010 levels using Bureau of Labor Statistics inflation rates for in-hospital care. Allareddy et al. Pediatric Emergency Visits for Facial Fracture. J Oral Maxillofac Surg with other facial bones. 5,6,12 It is likely that a substantial number of children have nasal bone fractures and are likely to be seen in outpatient clinical settings and not present to hospitalbased EDs. 12 Previous research has shown that motor vehicle accidents are the most common cause of facial fractures in children who require hospitalizations. 6-8,13,14 The present study showed that for children and adolescents presenting to the ED, the major causes of facial fracture were assaults (52.8% of ED visits) followed by falls (11.4%) and motor vehicle traffic accidents (10.9%). Of those who required hospitalization after an ED visit, motor vehicle traffic accident was the most common cause of injuries. It is likely that only patients with severe or multiple injuries required hospitalization, whereas those with less severe injuries were discharged from the ED without subsequent hospitalization. Findings from the present study also showed that age and gender of children and adolescents have a significant association with causes of injuries. Younger children were more likely to have falls, pedal cycle accidents, pedestrian accidents, and transport accidents, whereas older groups were more likely to have firearm injuries, motor vehicle traffic accidents, and assaults. Female patients were more likely to have falls, motor vehicle traffic accidents, and transport accidents, whereas male patients were more likely to have firearm injuries, pedal cycle accidents, and assaults. Those residing in low annual income household levels were at high risk for having firearm injuries, motor vehicle traffic accidents, and transport accidents compared with those residing in high income households. To the authors knowledge, none of the prior studies have examined the association between household income levels and causes of facial fracture injuries. The present study found that 5.3% of ED visits for facial fractures in this cohort occurred in children younger than 5 years, whereas 41.2% of ED visits occurred in those 11 to 17 years old and 45.6% occurred in those 18 to 21 years old. Close to 75% of ED visits concerned male patients. These findings are consistent with previous nationwide findings obtained from the Nationwide Inpatient Sample. 4 Vyas et al 4 examined facial fractures in children younger than 18 years and showed that children 1 to 4 years old comprised 5.6% of total hospitalizations, whereas children 15 to 17 years old comprised 50% of the cohort. 4 This study also found a significant male predomination, with a male-to-female ratio of 2.5:1. A significant amount of resources is used in the management of these fractures. Close to $929 million ED

8 ALLAREDDY ET AL 1763 Table 5. CHARACTERISTICS OF HOSPITALIZATION AFTER EMERGENCY DEPARTMENT VISITS (N = 44,627) Characteristic Hospitalizations, % Age (yr) Mean (SEM) 16.1 (0.2) Infants (<1 yr) 1.4 Toddlers (1-3 yr) 2.5 Preschool (4-5 yr) 2.7 School age (6-10 yr) 7.3 Early adolescents (11-14 yr) 10.1 Middle adolescents (15-17 yr) 23.3 Late adolescents (18-21 yr) 52.7 Gender Male 76.1 Female 23.9 Insurance Medicare 0.5 Medicaid 27.5 Private insurance 48.6 Uninsured 14.9 Other insurance 8.5 Types of facial fractures (ICD-9-CM codes) Closed fracture of nasal bones 28.8 (802.0) Open fracture of nasal bones 2.5 (802.1) Closed mandibular fracture, 31.5 combined Open mandibular fracture, 9.2 combined Closed fracture of unspecified 3.6 site of mandible (802.20) Closed fracture of condylar 4.3 process of mandible (802.21) Closed fracture of subcondylar 4.8 process of mandible (802.22) Closed fracture of coronoid 0.3 process of mandible (802.23) Closed fracture of unspecified 2.7 part of ramus of mandible (802.24) Closed fracture of angle of jaw 10.1 (802.25) Closed fracture of symphysis of 8.7 body of mandible (802.26) Closed fracture of alveolar 1.5 border of body of mandible (802.27) Closed fracture of other and 4.6 unspecified part of body of mandible (802.28) Closed fracture of multiple sites 2.4 of mandible (802.29) Open fracture of unspecified site 0.7 of mandible (802.30) Open fracture of condylar 0.4 process of mandible (802.31) Table 5. Cont d Characteristic Hospitalizations, % Open fracture of subcondylar 0.4 process of mandible (802.32) Open fracture of coronoid 0.09 process of mandible (802.33) Open fracture of unspecified 0.6 part of ramus of mandible (802.34) Open fracture of angle of jaw 2.9 (802.35) Open fracture of symphysis of 3.5 body of mandible (802.36) Open fracture of alveolar border 0.5 of body of mandible (802.37) Open fracture of body of other 2.1 and unspecified part of mandible (802.38) Open fracture of multiple sites 0.9 of mandible (802.39) Closed fracture of malar and 21.5 maxillary bones (802.4) Open fracture of malar and 2.0 maxillary bones (802.5) Closed fracture of orbital floor 19.4 (blowout) (802.6) Open fracture of orbital floor 1.0 (blowout) (802.7) Closed fracture of other facial 24.2 bones (802.8) Open fracture of other facial 1.6 bones (802.9) Year Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; SEM, standard error of the mean. Allareddy et al. Pediatric Emergency Visits for Facial Fracture. J Oral Maxillofac Surg charges and $3.07 billion total hospitalization charges were attributed to treating facial fractures in hospitalbased settings in the United States during the study period from 2008 to However, it should be kept in perspective that this study presents only hospital charges. Postdischarge costs, medication costs, and other indirect costs associated with hospitalizations are not included in the present study because of the lack of these data elements in the NEDS database. The present study has several strengths owing to the use of the largest all-payer discharge dataset in the United States. Considering the nationally representative nature of the dataset, the study results are

9 1764 PEDIATRIC EMERGENCY VISITS FOR FACIAL FRACTURE Table 6. HOSPITALIZATION OUTCOMES (DISPOSITION AFTER HOSPITALIZATION, LENGTH OF STAY, AND HOSPITALIZATION CHARGES) Hospitalizations, % Outcomes All Patients #14 yr old (n = 10,709) Patients yr old (n = 10,403) Patients yr old (n = 23,515) Disposition after hospitalization Routine Transfer to another shortterm hospital Transfer other (long-term care facilities, eg, skilled nursing facilities) Home health care Against medical advice 0.8 DS Died in hospital Discharge alive, destination 0.04 DS DS DS unknown/others Total charge for ED and inpatient services ($)* Mean (SEM) 69,103 (3,027) 53,542 (3,427) 70,707 (4,489) 75,512 (3,551) Total across entire United 3,070,367, ,690, ,636,943 1,765,040,331 States Length of stay (days) Mean (SEM) 5.0 (0.1) 4.3 (0.17) 4.9 (0.20) 5.3 (0.19) Total across entire United States 222,740 46,332 50, ,717 Abbreviations: DS, discharge information suppressed because of low cell counts (in accord with data user agreement with the Healthcare Cost and Utilization Project sponsored by the Agency for Healthcare Research and Quality); ED, emergency department; SEM, standard error of the mean. * All charges were adjusted for inflation to 2010 levels using Bureau of Labor Statistics inflation rates for in-hospital care. Allareddy et al. Pediatric Emergency Visits for Facial Fracture. J Oral Maxillofac Surg generalizable and externally valid. The authors provided estimates at a national scale and the detailed descriptive analysis identified high-risk cohorts who are likely to have facial fractures and seek hospitalbased settings for treatment. A comprehensive list of ICD-9-CM diagnostic codes was used to identify the different types of facial fracture and the external causes of injuries leading to these injuries. The authors hope that findings from the present study can help improve the assessment and management of these patients in hospital-based settings, leading to more efficient resource use and decreased financial burden on the health care system in the United States. The present study has limitations stemming from the use of large secondary hospital discharge datasets and the retrospective nature of the study design. The authors examined the association among age, gender, and annual household income levels on causes of fractures. They selected variables for the regression models based on a literature review. As mentioned in Materials and Methods, the NEDS dataset has many variables (close to 100). Most of these variables describe the diagnoses, external causes of injuries, occurrence of comorbid conditions, and outcomes (disposition after ED visits and subsequent hospitalization, charges, and length of stay). Most of these have been included in this study. The authors intentionally did not include urban versus rural designation in the model because this designation is applicable to the hospital and not to the individual patient. It would be interesting to examine zip-code level data of the patient and then correlate these with the hospital sought for treatment. Unfortunately, the NEDS dataset does not provide information on patient-level zip codes at this time. Furthermore, there is a lack of variables, such as behavioral variables, that would be of the greatest interest for injury-prevention programs. The retrospective nature of the study design precludes the authors from establishing a cause-and-effect relation. Facial fractures were identified using ICD-9-CM codes in the NEDS dataset. Any coding inconsistencies or coding errors at the time of data collection could yield biased estimates. As stated earlier, postdischarge outcome information is lacking in the NEDS database; hence, these were not examined.

10 ALLAREDDY ET AL 1765 During the 3-year period from 2008 to 2010, 336,124 ED visits were due to facial fractures in those no older than 21 years. Late adolescents, middle adolescents, and male patients comprised a significant proportion of these ED visits. Age, gender, and household income levels were significantly associated with the causes of facial fracture injuries. Press Release This article s Press Release can be found, in the online version, at References 1. Centers for Disease Control and Prevention: National estimates of 10 leading causes of non-fatal injuries treated in hospital emergency departments, United States Available at: Group_2010-a.pdf. Accessed December 30, Centers for Disease Control and Prevention: National estimates of 10 leading causes of non-fatal injuries treated in hospital emergency departments, United States Available at: In_Hospital%20Emergency_Dept_2011-a.pdf. Accessed December 30, Ryan ML, Thorson CM, Otero CA, et al: Pediatric facial trauma: A review of guidelines for assessment, evaluation, and management in the emergency department. J Craniofac Surg 22:1183, Vyas RM, Dickinson BP, Wasson KL, et al: Pediatric facial fractures: Current national incidence, distribution, and health care resource use. J Craniofac Surg 19:339, Siy RW, Brown RH, Koshy JC, et al: General management considerations in pediatric facial fractures. J Craniofac Surg 22:1190, Grunwaldt L, Smith DM, Zuckerbraun NS, et al: Pediatric facial fractures: Demographics, injury patterns, and associated injuries in 772 consecutive patients. Plast Reconstr Surg 128: 1263, MacIsaac ZM, Berhane H, Cray J Jr, et al: Nonfatal sport-related craniofacial fractures: Characteristics, mechanisms, and demographic data in the pediatric population. Plast Reconstr Surg 131:1339, Imahara SD, Hopper RA, Wang J, et al: Patterns and outcomes of pediatric facial fractures in the United States: A survey of the National Trauma Data Bank. J Am Coll Surg 207:710, HCUP Nationwide Emergency Department Sample (NEDS): Healthcare Cost and Utilization Project (HCUP) Rockville, MD, Agency for Healthcare Research and Quality. Available at: Hagan JF, Shaw JS, Duncan P (eds): Bright Futures: Guidelines for Health Supervision of Infants, Children, and Adolescents (ed 3). Pocket Guide. Elk Grove Village, IL, American Academy of Pediatrics, Bureau of Labor Statistics: Consumer Price Index inflation rate calculator for hospital inpatient care. Available at: bls.gov/data/inflation_calculator.htm. Accessed October 16, Cole P, Kaufman Y, Hollier LH Jr: Managing the pediatric facial fracture. Craniomaxillofac Trauma Reconstr 2:77, Eggensperger Wymann NM, H olzle A, Zachariou Z, et al: Pediatric craniofacial trauma. J Oral Maxillofac Surg 66:58, Haug RH, Foss J: Maxillofacial injuries in the pediatric patient. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 90:126, 2000