Firearm related injuries amongst children: Estimates from the nationwide
emergency department sample
Veerajalandhar Allareddya, Romesh P. Nalliahb, Sankeerth Rampac, Min Kyeong Kimd,
aDepartment of Pediatric Critical Care and Pharmacology, Rainbow Babies and Children’s Hospital, University Hospitals, Case Medical Center, United States
bSenior Tutor, Harvard School of Dental Medicine, United States
cNance College of Business Administration, Cleveland State University, OH, United States
dDMD student, Harvard School of Dental Medicine, United States
eDepartment of Developmental Biology Harvard School of Dental Medicine, 188 Longwoo Avenue, Boston, MA 02115, United States
Close to 50,000 persons die due to violence related injuries each
year in the United States.1Firearms are frequently used to commit
homicide and suicide.2Firearms are important sources of non-fatal
and fatal injuries amongst children in the United States and are
viewed as a public health problem.3–5It is important that policy
makers, providers, and public health authorities are kept abreast of
current estimates of firearm related injuries in children so that
preventive programmes and policies can be tailored to the needs of
cohorts that are deemed to be at the highest risk for experiencing
such injuries. Hospital based emergency departments (ED) are one
of the points of entry into the health care system for victims of
One national study of children up to age 14 has identified that
the major causes of firearm injuries are assault and accidental
injury by a known person.5However, there is a paucity of studies
providing estimates from hospital samples that are nationally
representative for children up to the age of 18. This represents an
important gap in knowledge since two previous studies found that
most of the paediatric firearm injuries occurred in the 15–19 age
The objective of the current retrospective descriptive study is
to provide estimates of firearm related injuries in children (up to
age 18) seeking care in hospital based emergency departments in
the United States. It is the authors’ hope that this study will
spark debate that leads to greater Federal Government regula-
tion of firearms in the United States. The Federal government
regulates medication (through the Food and Drug Administra-
tion) and consumer products (through the Consumer Products
Safety Commission), however, currently do not regulate fire-
arms. The importance of this study is to begin to use nationwide
Injury, Int. J. Care Injured 43 (2012) 2051–2054
A R T I C L E
I N F O
Accepted 29 October 2011
A B S T R A C T
Objective: The objective of this study is to provide estimates of firearm related injuries in children
seeking care in hospital based emergency departments.
Methods: The Nationwide Emergency Department Sample (NEDS) for the year 2008 was used for the
current study. All ED visits occurring amongst children aged less than or equal to 18 years and that had an
External Cause of Injury (E-Code) for any of the firearm related injuries were selected for analysis.
Results: A total of 14,831 ED visits (in children) in the United States had a firearm injury. The average age
of the ED visits was 15.9 years. Males constituted a predominant proportion of all ED visits (89.2%). A
total of 494 patients died in the emergency departments (3.4% of all ED visits) whilst 323 died following
in-patient admission into the same hospital (6% of all inpatient admissions). The most frequently
documented firearms were assaults by firearms and explosives (55% of all ED visits), accidents caused by
firearms and air gun missiles (33.6%), and injuries by firearms that were undetermined (7.4%). The
average charge for each ED visit was $3642 (25th percentile is $1146, median is $2003, and 75th
percentile is $4404). The mean charge for those visits that resulted in in-patient admission into the same
hospital was $70,164 (25th percentile is $16,704, median is $36,111, and 75th percentile is $74,165) and
the total charges for the entire United States was about $371.33 million.
Conclusions: The current study used the largest all-payer hospital based emergency department dataset to
provide national estimates of firearm related injuries amongst children in the United States during the year
2008 and highlights the public health impact of such injuries.
? 2011 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: +1 216 571 1009.
E-mail address: VA15@hsdm.harvard.edu (V. Allareddy).
Contents lists available at SciVerse ScienceDirect
jo ur n al ho m epag e: ww w.els evier .c om /lo cat e/inju r y
0020–1383/$ – see front matter ? 2011 Elsevier Ltd. All rights reserved.
estimates to document the impact on children that firearms are
having in the United States as measured through ED visits with
firearm related injuries.
Existing research has shown that 49% of unnatural deaths in
children are related to firearms8with the majority being homicide.
This indicates that although children are legally not permitted to
have access, firearm injury in children is an important public
health concern. One objective of our study is to provide
quantification of the firearm related injuries in children presenting
to ED’s across the United States.
Materials and methods
The Nationwide Emergency Department Sample (NEDS) for
the year 2008 was used for the current study. The NEDS dataset is
sponsored by the Agency for Healthcare Research and Quality
(AHRQ) and is a component of Healthcare Cost and Utilisation
Project.9The NEDS database is a 20% stratified sample of hospital
based emergency department (ED) visits in the United States
and is the largest all-payer ED database that has been designed
to enable examine hospital based emergency department out-
comes and utilisation patterns across the United States.9The
NEDS 2008 database draws ED visit information from twenty
eight participating states. The sample weights available in the
database can be employed to calculate national estimates of all
hospital based ED visits. NEDS provides information on more
than 100 variables including patient characteristics (including
age, gender, diagnoses documented during the ED visit and
subsequent hospitalization, procedures performed during hos-
pitalization, disposition information from ED and hospitaliza-
tion, insurance status, and hospital charges) and several hospital
Data user agreement
Since the current study was a secondary data analysis of
NEDS dataset made available by AHRQ, institutional review
board approval was not required. The first author (VjA) obtained
data from AHRQ after completing the data user agreement with
HCUP. Since the HCUP data user agreement precludes reporting
individual cell counts of less than or equal to 10 to preserve
patient confidentiality, these numbers were not reported in the
current study. The term ‘‘DS’’ was used in tables wherever
individual cell counts were less than or equal to 10.
All ED visits occurring amongst children aged less than or equal
to 18 years and that had an External Cause of Injury (E-Code) for
any of the firearm related injuries (see Table 2 for list of E-codes
used) were selected for analysis. The NEDS database has four fields
to capture E-Codes and ICD-9-CM codes mapping to fire arm
injuries were used to identify cases. The NEDS database also has 15
diagnoses fields that enable researchers to identify the reasons for
ED visits or hospitalization. Injuries to different body parts were
documented by using ICD-9-CM codes in these diagnoses fields.
Information regarding hospital charges and length of stay and
other variables examined in this study were obtained from the
variables in the NEDS database. Sample weights assigned to each
discharge were used to project all estimates to national levels.
Descriptive statistics were used to summarize the data. For all
analyses, the NEDS hospital stratum was the stratification unit and
each individual ED visit was the unit of analysis. All statistical
analyses were conducted using SAS (Version 9.2-SAS Institute,
In the year 2008, a total of 14,831 paediatric ED visits in the
United States had a firearm injury. The average age of the ED visits
was 15.9 years. Close to 39% of all ED visits due to firearm injuries
occurred during the weekends (Table 1). Males constituted a
predominant proportion of all ED visits (89.2%). With regards to
disposition of patient following an ED visit, 51.5% were discharged
routinely whilst 5.1% were transferred to another short term
hospital, 0.7% were transferred to other facilities (including skilled
nursing facility and intermediate care facilities), and 0.5% were
discharged against medical advice. A total of 494 patients died in
the emergency departments (3.4% of all ED visits). Admission into
the same hospital was required for 5342 ED visits. Amongst ED
visits who were admitted into the same hospital, 77.9% were
routinely discharged, 3% were transferred to another short term
hospital, 5.4% were transferred to ‘‘other’’ facilities, 4.8% dis-
charged to home health care, and 0.9% were discharged against
medical advice. A total of 323 patients died in the same hospital
following an ED visit (6% of all inpatient admissions). About 52.9%
of all ED visits where from patients residing in geographic areas
with a median household income level of less than $39,000.
Medicaid was listed as the primary payer for 46.1% of all ED visits
whilst 22.2% of the ED visits were not covered by any insurance
plans. Emergency departments attached to metropolitan teaching
hospitals treated most visits (65.7%).
Sources of firearms used to injure subjects were found by using
E-codes and are summarized in Table 2. The most frequently
documented firearms are assaults by firearms and explosives (55%
of all ED visits), accidents caused by firearms and air gun missile
(33.6%), injuries by firearms/air guns/explosives – undetermined
whether accidentally or purposely inflicted (7.4%), suicide and self
inflicted injuries (2.7%), and injuries due to legal intervention by
Injuries resulting from use of firearms are presented in Table 3.
Open wounds of extremities were documented in 49.9% of all ED
visits. Other frequently documented injuries include open wounds
of head, neck, and trunk (27.7%), crushing or internal injuries
(18.3%), fracture of lower limbs (10.2%), and fracture of upper limbs
(8.9%). Intra cranial injuries were documented in 4.8% of ED visits.
The average charge for each ED visit was $3642 (25th percentile
is $1146, median is $2003, and 75th percentile is $4404). The total
ED charges for the entire United States was $38.21 million. The
mean charge for those visits that resulted in in-patient admission
into the same hospital was $70,164 (25th percentile is $16,704,
median is $36,111, and 75th percentile is $74,165) and the total
charges for the entire United States was about $371.33 million. The
mean length of stay in hospital was 6.25 days the total
hospitalization days in the United States was 33,404 days.
Our study provides estimates of hospital based emergency
department visits attributed to firearm injuries in children aged 18
years and younger in the United States during the year 2008.
Results from the current study indicate that 14,831 ED visits
occurred in the year 2008 and a total of 817 children died either in
the emergency department or during subsequent admission into
the same hospital. A substantial amount of hospital resources are
utilized to treat these patients. The current study highlights the
public health impact of firearm injuries in a population cohort that
is not legally permitted to have access to firearms which
underlines the need to enforce and strengthen laws pertaining
to firearms access. Improving enforcement and reducing access to
firearms could result in a reduction in ED visits and mortality rates
due to firearm injuries.
V. Allareddy et al. / Injury, Int. J. Care Injured 43 (2012) 2051–2054
The current study demonstrates that more than half of the ED
visits due to firearm injuries occurred in regions with a median
household income below $39,000. Almost 78% of these visits were
in regions with a median household income below $49,000. In
2008 (the same year as our study), the median household income
in the Unites states was $52,02910which indicates that 78% of ED
visits due to firearm injuries occurred in regions with a household
income below the national median household income. Educational
programmes for parents who reside in these regions about firearm
safety could result in a large reduction in ED visits and mortality
rates due to firearm injuries.
A disproportionate proportion of ED visits due to firearm
injuries occurred on Saturday and Sunday. The weekend represents
about 29% of the week but produced 40% of the ED visits for firearm
injuries. Previous research found that 65% of paediatric firearm
injuries occurred between 5pm and 5am.6The development of
structured non-school hours entertainment, education or athletic
programmes for children may reduce the incidence of ED visits due
to firearm injuries.
Previous research in children under the age of 14 has shown
that about 40% of firearm injuries were accidental5and 41% were
the result of assaults. Our study also found accidental firearm
injuries to be close to 33.6%, but found assault related injuries to be
about 55% (14% more than the previous study). Our study also
considered children aged 14–18 and the difference in the study
sample may be attributable to the higher incidence of firearm
assault in our study.
From a public health perspective, it is critical to identify the
cohort of population that is at a high risk to experience such
injuries and target educational and social interventional strategies
Types of fire arm injuries reported.
Source of fire arm injury (ICD-9-CM Code)
95% C.I. of Estimate
Accident caused by firearm and air gun missile (E922)
Suicide and self-inflicted injury (E955)
Assault by firearms and explosives (E965)
Injury due to legal intervention by firearms–including: Gunshot wound machine gun,
revolver, rifle pellet or rubber bullet, not otherwise specified (E970)
Injury by firearms, air guns and explosives, undetermined whether accidentally or purposely inflicted (E985)
Note. Individual cell counts may not add to the global total of 14,831 discharges since a single discharge may have more than one type of external cause of injury coded in the
Characteristics of Hospital Based Emergency Department visits with fire arm injuries amongst children.
95% C.I. of estimate
Admission on Monday–Friday
Admission on Saturday–Sunday
Disposition of patient from the ED
Transfer to short term facility
Other transfers, including skilled nursing facility,
intermediate care, and another type of facility
Home health care
Against medical advice
Admitted as an inpatient to this hospital
Died in ED
Not admitted, destination unknown
Median household income quartiles for patient’s ZIP Code
$64,000 or more
Private including HMO
Hospital location and teaching status
Metropolitan non teaching
DS, ‘‘Discharge Information Suppressed’’ since cell counts were less than or equal to 10 (as per data user agreement with AHRQ).
Note. Individual cell counts may not add to the global total of 14,831 discharges because of missing values.
V. Allareddy et al. / Injury, Int. J. Care Injured 43 (2012) 2051–2054
towards these cohorts.11,12The current study results indicate that
a greater proportion of the ED visits occurred amongst males and
those covered by Medicaid. However, the nature of the dataset
precludes us from conclusively proving a cause and effect
relationship between these demographic characteristics and risk
of firearm injuries either accidentally or through assaults. More
research is necessary to identify perpetrators of assaults in such
high risk populations. Unfortunately, the NEDS database does not
capture this information. Previous studies have shown that lower
socioeconomic status Latino and African American youth are at a
higher risk for experiencing firearm related injuries.13Race
information is not available in the NEDS dataset and hence we
are unable to examine this association.
The current study has demonstrated 14,831 ED visits by
children due to firearm injury occurred in year 2008. Further, 817
of these children died. This is an alarming finding when it is taken
into consideration that children may not legally have access to
firearms. Making firearm access laws more strict may help to
reduce death from firearm injury in children. Previous research has
shown that 18% of parents with firearms, keep their firearm within
reach of their child14and 26% of parents keep the firearm loaded.
These are extremely concerning statistics and federal funded
programmes to educate parents of child firearm safety may be a
crucial way to reduce the incidence on injury and mortality due to
The results presented in this retrospective study of secondary
hospital discharge datasets should be interpreted in the perspec-
tive of several limitations of the dataset used. As mentioned in the
methods section, type of firearm related injuries were identified by
using external cause of injury ICD-9-CM codes. Consequently the
estimates presented in the current study are not likely to be
accurate as there could be variations in coding practises of
hospitals. The resource utilisation in terms of hospital charges
presented in the current study is an underestimation of the actual
economic costs associated with firearm injuries as out of hospital
costs and in-direct costs associated with these injuries is not
captured in the dataset. Key information that would aid in
developing injury prevention programmes including location of
injuries, the precipitating events or circumstances such as disputes
leading to firearm injuries are not captured in the current dataset.
Data regarding lab testing (drug or toxicology reports) are also not
available in the NEDS dataset. Estimates presented in the current
study are likely to be an underrepresentation of the true firearm
injury related event rates amongst kids since injured individuals
are also likely to be treated in physician offices, community clinics,
or might have died even before reaching the hospital based
emergency departments. Finally, there is some missing informa-
tion regarding certain demographic attributes presented in Table 1.
Only three variables had missing information. Gender information
was not available for 11 ED visits (0.07% of all ED visits), insurance
status information was not available for 54 ED visits (0.36% of all
ED visits), and median household income level information was
not available for 465 visits (3.1% of all ED visits). Considering the
unique and large number of cases in our dataset, we believe that
the amount of missing information is small and will not adversely
bias our estimates.
Despite the above mentioned weaknesses, the current study
also has several strengths. The NEDS is the largest all-payer dataset
in the United States capturing discharge and disposition informa-
tion on all ED visits. The dataset allows us to project to national
estimates and thus the study results are more generalisable
compared to those originating from single centre studies.
The current study used the largest all-payer hospital based
emergency department dataset to provide national estimates of
firearm related injuries amongst children in the United States
during the year 2008. More than 50% of such visits were attributed
to assaults. A greater proportion of such ED visits appear to be
occurring amongst males, those residing in low income level areas,
and those covered by Medicaid insurance programme (except
‘program’ in computers). The current study results also highlight
the adverse outcomes associated with such injuries including 817
deaths amongst kids and substantial hospitalization charges
1. Katch DL, Dahlberg LL, Patel N. Morbidity and Mortality Weekly Report –
Surveillance for violent deaths – National Violent Death Reporting System,
16 States, 2007. Department of Health and Human Services – Centers for
Disease Control and Prevention. May 14, 2010. Volume 59. No. SS-4.
2. Byard RW, Haas E, Marshall DT, Gilbert JD, Krous HF. Characteristic features of
pediatric firearm fatalities—comparisons between Australia and the United
States. J Forensic Sci 2009;54(5):1093–6.
3. Powell EC, Jovtis E, Tanz RR. Incidence and circumstances of nonfatal firearm-
related injuries among children and adolescents. Arch Pediatr Adolesc Med
4. Hemenway D. The public health approach to motor vehicles, tobacco, and
alcohol, with applications to firearms policy. J Public Health Policy 2001;22:
5. Eber GB, Annest JL, Mercy JA, Ryan GW. Nonfatal and fatal firearms-related
injuries among children aged 14 years and younger: United States, 1993–2000.
6. Ary RD, Waldrop RD, Harper DE. The increasing burden of pediatric firearm
injuries on the emergency department. Pediatric Emerg Care 1996;12:391–3.
7. Zavoski RW, Lapidus GD, Lerer TJ, Banco LI. A population-based study of
severe firearm injury among children and youth. Pediatrics 1995 Aug;96(2 Pt 1):
8. Heninger M, Hanzlick R. Nonnatural deaths of adolescents and teenagers:
Fulton County Georgia, 1985–2004. Am J Forensic Med Pathol 2008 Sep;29(3):
9. AHRQ. Agency for Healthcare Research and Quality. Nationwide ED Sample
Database Documentation – Year 2007 Documentation. Obtained from http://
www.hcup-us.ahrq.gov/db/nation/neds/nedsdde.jsp [accessed 22.12.10].
10. US Census Bureau. American Community Survey Reports – Year 2008. Obtained
from http://www.census.gov/prod/2009pubs/acsbr08-2.pdf [accessed 30.09.11].
11. Coyne-Beasley T, Schoenbach VJ, Johnson RN. Love our kids, lock you guns—a
community-based firearm safety counseling and gun lock distribution pro-
gram. Arch Pediatr Adolesc Med 2001;155:660–4.
12. Andres AR, Hempstead K. Gun control and suicide: The impact of state firearm
regulations in the United States, 1995–2004. Health Policy 2010. [October 31
(Epub-ahead of print)].
13. Powell EC, Tanz RR. Child and adolescent injury and death from urban firearm
assault: association with age, race, and poverty. Injury Prev 1999;5:41–7.
14. Knight-Bohnhoff K, Harris MB. Parent’s behaviors, knowledge, and beliefs
related to unintentional firearm injuries among children and youth in the
southwest. J Pediatr Health Care 1998;12(May–June (3)):139–46.
Type of injuries documented in the dataset.
Type of injury
95% C.I. of
Joint disorders and dislocations;
Fracture of neck of femur (hip)
Spinal cord injury
Skull and face fractures
Fracture of upper limb
Fracture of lower limb
Sprains and strains
Crushing injury or internal injury
Open wounds of head, neck, and trunk
Open wounds of extremities
DS, ‘‘Discharge Information Suppressed’’ since cell counts were less than or equal to
10 (as per data user agreement with AHRQ).
Note. Individual cell counts may not add to the global total of 14,831 discharges
since a single discharge may have more than one type of injury coded in the dataset.
V. Allareddy et al. / Injury, Int. J. Care Injured 43 (2012) 2051–2054