METHODS

We used International Classification of Diseases, 10th Revision (ICD-10) discharge diagnosis codes to conduct a cross-sectional comparison of pediatric (age 0–18 years) firearm-related and total emergency department (ED) and hospital encounters between calendar weeks 9 and 36 (corresponding to approximately March to August) 2020 and 2017–2019 at 44 US children’s hospitals participating in the Pediatric Health Information System (PHIS) database. The primary outcome was firearm-related encounters, defined as an initial encounter for a penetrating injury from a powder-charged weapon identified by ICD-10 discharge code (see Supplemental Information).⁶ The median weekly volumes in 4-week blocks for 2017–2019 were compared to the median in 2020 by using the Wilcoxon rank test. Differences in medians were estimated by using quantile regression. Similarly, weekly volumes in 2020 were compared to the 25th and 75th percentile for 2017–2019.

RESULTS

There were 2510 firearm-related encounters during the 4-year study period, of which 798 (32%) occurred in 2020. Of these, 45.6% included adolescents aged 15 to 18 years, 78.8% were boys, 61.7% were non-Hispanic Black children, 62.7% lived in the South census region, and 48.9% resided in a very low Child Opportunity Index neighborhood (Table 1). There were no differences in cohort characteristics in 2020 vs 2017–2019 (Table 1).

TABLE 1

Characteristics of Study Population, March to August 2017–2020

	Firearm-Related Encounters
Characteristics	2017–2019 (n = 1712), n (%)	2020 (n = 798), n (%)	P
Age, y
0–4	168 (9.8)	94 (11.8)	.381
5–9	209 (12.2)	105 (13.2)
10–14	516 (30.1)	235 (29.4)
15–18	819 (47.8)	364 (45.6)
Sex
Male	1339 (79)	622 (78.8)	.905
Female	355 (21)	167 (21.2)
Race and/or ethnicity
Non-Hispanic white	313 (18.3)	138 (17.3)	.185
Non-Hispanic Black	1064 (62.1)	492 (61.7)
Hispanic	222 (13)	103 (12.9)
Asian	5 (0.3)	8 (1)
Other	108 (6.3)	57 (7.1)
Payer
Government	1286 (75.1)	621 (77.8)	.135
Private	285 (16.6)	108 (13.5)
Other	141 (8.2)	69 (8.6)
Intention
Unintentional	988 (57.7)	502 (62.9)	.053
Intentional or self-harm	46 (2.7)	17 (2.1)
Assault	593 (34.6)	243 (30.5)
Undetermined	58 (3.4)	25 (3.1)
Legal intervention	5 (0.3)	6 (0.8)
Multiplea	22 (1.3)	5 (0.6)
Census region
Midwest	454 (26.5)	192 (24.1)	.330
Northeast	99 (5.8)	40 (5)
South	1007 (58.8)	500 (62.7)
West	152 (8.9)	66 (8.3)
Rurality
Rural	171 (10)	77 (9.6)	.616
Urban	1516 (88.6)	713 (89.3)
Child Opportunity Indexb
Very low	879 (51.3)	390 (48.9)	.339
Low	354 (20.7)	178 (22.3)
Moderate	232 (13.6)	121 (15.2)
High	155 (9.1)	64 (8)
Very high	87 (5.1)	39 (4.9)
Disposition
Died	95.5 (5.5)	50 (6.3)	.659
ED treat and/or release	689 (40.2)	328 (41.1)
Hospitalized	928 (54.2)	420 (52.6)

Open in a separate window

^aMultiple intent category used for cases with >1 assigned ICD-10 discharge diagnosis code

^bThe Child Opportunity Index is a composite metric based on 29 neighborhood indicators collected from public and proprietary sources at the census-tract level, assigned to 5 ordered levels of opportunity (very low, low, moderate, high, and very high) to describe the level of neighborhood opportunity.¹²

Compared to 2017–2019, total encounters in 2020 decreased in each 4-week block and by 42.3% overall (Fig 1A). In contrast, firearm-related encounters increased by 38.8% from a median of 575 encounters in 2017–2019 to 798 encounters in 2020, increasing in each 4-week block with a range of 116.9% to 167.9% (Fig 1B). An observed increase in overall firearm-related encounters in 2020 remained when using 25th percentile (+57.7%) or 75th percentile (+26.5%) volumes for 2017–2019. The overall difference in medians was statistically significant (+9.0 encounters per week; P < .001).

Open in a separate window

FIGURE 1

Comparing total and firearm-related encounters in 2020 to 2017–2019. A, Total ED and hospital encounter volumes in 2020 compared to 2017–2019. B, Firearm-related encounter volumes in 2020 compared to 2017–2019.

DISCUSSION

Firearm-related encounters at US children’s hospitals increased significantly in March to August 2020 compared to 2017–2019, whereas total encounters decreased substantially.

The combination of unparalleled firearm sales, heightened social and economic distress, and children atypically home from school created a uniquely risky environment for firearm injury. Consistent with previous literature, we found higher rates of firearm injuries in older non-Hispanic Black adolescent boys from Southern regions.⁷ In addition, we found no significant changes in demographics, rurality, or region comparing firearm-related encounters in 2020 to 2017–2019, suggesting the SARS-CoV-2 pandemic was associated with an increase in pediatric firearm injuries without a shift in the population most at risk. Additionally, nearly half of our firearm cohort resided in a very low Child Opportunity Index neighborhood. This highlights socioeconomic inequities, many of which overlap with risk factors associated with violence, including poverty, inadequate housing, and lower economic opportunity.^8,9 Although in recent studies, researchers also found that firearm injuries among adults and very young children increased during the SARS-CoV-2 pandemic, this is the first report highlighting the significant increase in pediatric firearm encounters for all ages during the pandemic despite unprecedented declines in total encounters.

Limitations include retrospective design, restricted time frame of available data, and reduced generalizability to community hospitals and rural settings. Despite likely underrepresenting older adolescents managed in adult trauma centers, we demonstrated a significant increase in firearm encounters, highlighting the impact firearm injuries have on children of all ages. Although we recognize database limitations, we chose PHIS given its nationally representative sample, inclusion of inpatient and ED encounters, relatively even representation of injury intents, and up-to-date data. Lastly, we did not make inferences on the basis of coded injury intent because of known discordance between codes and injury description.¹⁰ Different trends may emerge as extended time lines and additional data sets are used.

These findings underscore the critical importance of preventing firearm injuries, intensified in the context of a global pandemic, by applying effective tools such as counseling on safe firearm storage, promoting crisis support and violence intervention resources, supporting responsible firearm legislation, and advocating for firearm safety research.¹¹

Supplementary Material

FUNDING:

ABBREVIATIONS

Footnotes

REFERENCES