Injury-Related Pediatric Emergency Department Visits in the First Year of COVID-19

We conducted a cross-sectional study using data from the Pediatric Health Information System (PHIS), which includes 49 tertiary care children’s hospitals in the United States. The PHIS database contains International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) diagnosis and procedure codes, billed services, and hospital charges. Data integrity is jointly monitored by participating institutions and the Children’s Hospital Association.¹¹  We included all ED visits for children aged 0 to 17 years during the first year of the SARS-CoV-2 pandemic (March 15, 2020 to March 14, 2021) and a 3 year comparator period (March 15–March 14, 2017–2020) for 41 hospitals with complete administrative and billing data for the study periods. Encounters for children born at the hospital, those who left without being seen by an ED provider, and those who left against medical advice were also excluded. The Nationwide Children’s Hospital institutional review board deemed that this study did not qualify as human subject research.

We divided the first year of the pandemic into periods on the basis of widely experienced changes in social distancing, and defined the early pandemic period as the first Sunday (March 15, 2020) after the national emergency declaration on March 13, 2020 to June 30, 2020.¹²  The middle (July 1, 2020 to October 31, 2020) and late (November 1, 2020 to March 14, 2021) pandemic periods were selected specifically to account for trauma seasonality, because trauma visits typically occur with the highest incidence between April and late October, and to reflect the return to in-person or hybrid models for school attendance for many children in the late period.¹³  ED visit numbers for the 3 year comparator period were averaged across the same calendar dates as the pandemic period to consider yearly variation in ED volume and case mix.

Injury encounter identification was performed using external causes of injury codes (specifically, V, X, W, and Y codes in ICD-10-CM coding). The sample was restricted to encounters with external cause of injury codes, categorized using the ICD-10-CM groupings in the Centers for Disease Control and Prevention’s external cause of injury matrix, for primary injury mechanisms of bites/stings, cut/pierce, drowning, environmental, fall, fire/burns, firearm, foreign body, motor vehicle collisions, neglect/abuse, overexertion, pedal cyclist, pedestrian, poisoning, sports-related injury, struck by/against, suffocation/asphyxiation, and transportation included.¹⁴ 

Demographic characteristics for each ED visit included age (<1, 1–4, 5–9, 10–14, and 15–17 years), sex, race and ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, Asian American or other), payer (government, private, or other), distance from the hospital (calculated as the number of miles between the centroids of the patient’s and hospital’s zip codes), geographic region (Midwest, Northeast, South, West), rurality (metropolitan, micropolitan, small town, rural), and median zip code income.¹⁵  Race and ethnicity data were preserved in the analyses as a social construct because of their association with differences in ED care-seeking behavior and in injury rates.^16–18  The exposure of interest was documented race and ethnicity. In the PHIS, race and ethnicity are included as 2 distinct variables, which were collapsed into a single variable.¹⁹  Hospitals submit race and ethnicity data to the PHIS for each visit according to hospital-specific practices, which include parent/guardian self-report at the time of arrival or hospital registration assignment. Zip code-level social determinants of health were measured using the Child Opportunity Index 2.0, which includes 29 indicators across 3 domains of neighborhood characteristics that impact children’s healthy development: education, health and environment, and social and economic opportunities.²⁰  Child opportunity categories are created by ranking all US zip codes by overall score and dividing them into equally sized quintiles by population: very low, low, moderate, high, and very high.

Measures to evaluate injury type and severity included reported intentionality, injury severity score (ISS), and ED disposition.²¹  We categorized injury severity by ISS as follows: minimal, (0), mild, (1 to 8), moderate, (9 to 15), and serious–critical (≥16).²²  Finally, we categorized ED injury visits by both injury patterns (superficial, open wound, sprains/strain, fracture, and dislocation) and injury regions (head, neck, face, thorax, abdomen, pelvis, upper and lower extremity, and other). We examined primary discharge diagnoses, ED management, and disposition for each visit.²³ 

Patient demographic and clinical characteristics were summarized as frequencies for categorical variables and median with interquartile ranges for numerical variables. To calculate percentage change between 2019 (comparator) and 2020, the number of clinical encounters in a given 2019 interval was subtracted from the number of encounters during the corresponding 2020 interval and then divided by the number of encounters in the 2019 interval, which we adapted to our multiple-year comparison periods. Percentage changes in absolute visits were calculated by season in total and by subgroups. Percentage change by ISS in total was also stratified by race/ethnicity, Child Opportunity Index (COI), median zip code income, and rurality. Because of the large sample size, formal statistical comparisons were not made, but an a priori absolute difference of 5% among groups was considered clinically significant.²⁴  SAS 9.4 (SAS Institute, Inc, Cary, NC) was used for all analyses.

Between the pandemic study period and the 3 comparator years, we identified 2 846 991 total pediatric injury visits (Fig 1). Compared with the 3 previous years, demographic characteristics of the injured children presenting to the ED differed during each of the pandemic periods (Table 1). ED injury-related ED visits demonstrated a sharp decline during the early pandemic period (34.4%), followed by modest declines in volume during the middle (24.4%) and late (21.2%) pandemic periods. There were sharper declines among school-aged children and adolescents compared with younger children, and among non-Hispanic Black, Hispanic, and Asian American patients compared with non-Hispanic White patients. The decline in injury-related ED visits was greatest for metropolitan communities and those in closer proximity to children’s hospitals. Among both COI quintiles and median household income quartiles, the number of visits declined most in the very-low opportunity and the lowest-income groups, with a gradated exposure–response relationship through the remaining COI and income groups.

By injury intent, ED visits decreased during each pandemic period except for suicide intent, which increased during the middle (19.9%) and late (21.8%) pandemic periods (Table 2). By injury severity, ED injury-related visits showed an increase in serious–critical injuries that began in the early period (15.9%) and continued throughout the pandemic (4.9% middle; 20.6% late). Consistent with the observed increase in children with severe injury, injury-related visits resulting in discharge from the ED declined during all pandemic periods (−36.9% early; −26.9% middle, −24.3% late), whereas visits resulting in hospital admission (−13%; −1.5%; 3%) or ICU admission (−10.3%; 1.7%; 4.1%) trended upward through the study period. A marked increase in deceased patients who arrived in the ED was observed in the early (37.1%), middle (43.6%), and late (14%) pandemic periods.

Compared with baseline, during all 3 pandemic periods, the number of visits by mechanism category declined with the following exceptions: there were increases in injury-related ED visits because of firearms (22.9% early; 42.8% middle; 37% late), pedal cyclist (60.4%.; 24.9%; 32.2%), other transportation (20.8%; 25.3%; 17.9%), and suffocation/asphyxiation (21.4%; 20.2%; 28.4%). The number of visits because of cutting/piercing increased in the late pandemic period only (7.2%) (Table 3). Although declines in ED visits for all patterns of injury were observed, the sharpest declines were for superficial injuries (early −41.7%) and sprains/strains (early –62.4%) (Table 4). Table 5 describes the distribution of severe and nonsevere injuries by intent and mechanism. Most severe injuries were unintentional (61%), and the leading mechanisms for severe injury observed in children remained motor vehicle crash–occupant (31.4%), sports-related injury (15.8%), fall (12.7%), and transportation–other (10.2%). The distribution of severe injuries before the SARS-CoV-2 pandemic and during the SARS-CoV-2 pandemic remained largely unchanged.

Exploratory analyses evaluated the relationship between injury severity and COI, race, ethnicity, household income, and rurality (Fig 2). ED visits for minimal, mild, and moderate severity injuries decreased across all COI, race, ethnicity, income, and urbanicity groups. Declines were greatest among children with very-low COI, living in zip codes associated with the lowest-income quartile and/or in metropolitan areas, and those who were Black or Hispanic. Serious–critical injuries, however, increased across all COI, race, ethnicity, income, and urbanicity groups during the pandemic. The greatest increases in serious–critical injuries were observed among children with high or very-high COI, those living in zip codes associated with the highest 3 income quartiles and/or rural areas, and those who were Black or Asian American.

Overall pediatric injury-related ED encounters at US children’s hospitals decreased substantially during the pandemic; however, relative injury severity increased. This study continues to expand the knowledge of the impact of the pandemic on pediatric injury events and, to our knowledge, is 1 of the earliest to evaluate national trends in pediatric injury-related ED encounters at children’s hospitals during the first year of the SARS-CoV-2 pandemic. In addition to the previously known decline in pediatric ED visits during the SARS-CoV-2 pandemic, this study found an increase in severity of injuries including deaths, which continued to increase as the pandemic progressed. The overall distribution of injuries before the SARS-CoV-2 pandemic and during the SARS-CoV-2 pandemic were similar, but there was an increase in injury mechanisms that are disproportionately severe. For example, injuries with the associated mechanisms for firearm, pedal cyclist, and suffocation/asphyxiation increased, and injuries associated with suicide intent were increased throughout the pandemic period, which are injuries that are more likely to be severe. Nearly all of the suffocation/asphyxiation injuries were also designated also as intentional self-harm (data not shown). The degree of decline in overall ED injury-related visits and the severity of injury events during the first year of the pandemic appeared to be related to child opportunity; lower opportunity related to greater decreases in overall injury-related ED visits and higher opportunity related to greater increases in severity of injury events.

Although it is unclear at this time why the conditions of the SARS-CoV-2 pandemic were associated with increases in pediatric injury severity, previous studies have demonstrated severe pediatric traumatic injuries occur at higher rates during the spring, summer, afternoons, evenings, and weekends.^25,26  Without the inherent structure of school and extracurricular activities, children may have been at increased risk for severe injury while engaging in unsupervised play/activity. Although a single center found an association with reduced pediatric trauma visits with reduced school days during the pandemic, there is a need to also consider the impact of risk-taking behaviors relative to school closures.²⁷  Other literature supports the association of risky behaviors with unsupervised, out of school time, which likely increased during the SARS-CoV-2 pandemic.²⁸  Understanding the changes in patterns and mechanisms of pediatric injuries relative to changes in family and school structure during the pandemic could help guide the development of injury prevention strategies, treatment, and management.

Serious–critical injuries increased in all categories of child opportunity, race and ethnicity, median zip income, and urbanicity. Children from rural areas and Black, Hispanic, and Asian American children had larger increases in ED visits for severe injury. Increased risk-taking behavior among rural youth and decreased adherence to injury prevention practices in rural areas are well documented in the literature and may account for the observed increases in ED visits for severe injury.^29,30  In addition, there is greater access and use of motorized equipment, along with dangerous roads with high speeds in rural areas, thus increasing the risk for severe injuries.³¹  In the setting of the SARS-CoV-2 pandemic, the already high and rising suicide rates among children living in rural area and Asian American and Black children may have been exacerbated.^32–34  The dramatic change in family and school structure during the SARS-CoV-2 pandemic may have compounded existing differences in injury trends among children living in rural areas and for Black, Hispanic, and Asian American children. Our study found relatively large decreases in ED injury visits for children in the lower socioeconomic status and Black, Hispanic, and Asian American children. Consistent with previous literature, 1 explanation for these decreases may be lower ED utilization from historically marginalized groups during the pandemic amid social distancing measures in place.^35,36  Additionally, it may be that Black, Hispanic, and Asian American children, especially in urban environments, had greater decreases in their injury rates because of closer supervision by parents who remained home during the pandemic because of joblessness or transitions to a new work-from-home status.^37,38 

A previous, single-center, pandemic-related report also showed an increase in penetrating trauma during the pandemic period, which was offset by a decline in motor vehicle-related injuries, so that the absolute number of severe injuries remained constant.¹⁰  Consistent with previous studies, we redemonstrate an increase in pediatric firearm-related injuries during the SARS-CoV-2 pandemic.^39,40  The SARS-CoV-2 pandemic brought about record sales of new firearms.⁴¹  Together with the demonstrated increase in pediatric suicide intent, these sales spark significant concern because of the added psychological and financial stressors of theSARS-CoV-2 pandemic, which are known risk factors for suicide attempts and death.^42,43  Social distancing as a mitigation effort to reduce the virus transmission has been implicated as a new psychological stressor impacting pediatric suicide attempts and deaths during the SARS-CoV-2 pandemic.⁴⁴  The increase in both injuries with suicide intent and firearm injuries underscores the importance of mental health assessments, social supports and resources, and firearm harm reduction strategies during pandemics and their aftermath.^42,45 

Pedal cycling and other ride-on apparatuses (eg, scooters) offered means of transportation and leisure time activity for children during SARS-CoV-2 stay-at-home orders amid school closures and cancellations of regular sporting activities. Consistent with another pediatric study, there was an increase in pediatric pedal cyclist and other transportation injuries throughout all pandemic periods.⁴⁶  The increase in pediatric bicycle-related injuries could also be attributed to increased bicycle sales and use of bicycle share programs during the same time period.⁴⁷  Although our results showed an increase in moderate to severe–critical injury visits in all opportunity groups, it was larger in the high and very-high opportunity groups that may have greater access to these means of transportation under pandemic conditions.⁴⁷ 

Limitations include retrospective design and reduced generalizability to community hospitals and rural settings. Without access to the medical records in our database, we were unable to ascertain the circumstances that surrounded the injuries. Although we recognize database limitations, we chose PHIS given its large national sample, inclusion of inpatient and ED encounters, and up-to-date data. Although we reported intent on the basis of ICD-10 coding, we refrained from inferences because of known discordance between codes and injury description.^48,49  We recognize that decreased ED visits for minor injuries does not necessarily negate their existence but may be attributable to avoidance of SARS-CoV-2 exposure and other nonclinical factors. Different injury trends may continue to emerge as extended timelines and additional data sets are used.

Pediatric EDs in the United States saw a dramatic decline in injury-related visits during the SARS-CoV-2 pandemic. Although visits declined across a broad range of injury types, the proportional increase in injury severity, increases in pedal cyclist, firearm, and suffocation/asphyxiation injuries, as well as injuries because of suicide intent, are concerning. As public health officials continue to urge preparation for ongoing waves of the SARS-CoV-2 pandemic, as well as future pandemics, it is of critical importance to prioritize research and develop strategies that promote injury prevention.

COI

Child Opportunity Index

ED

emergency department

ICD-10-CM

International Classification of Diseases, 10^th Revision, Clinical Modification

ISS

injury severity score

PHIS

Pediatric Health Information System

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2