Article Text


Epidemiology of paediatric firearm injuries in the USA, 2001–2010
  1. Saranya Srinivasan1,
  2. Rebekah Mannix2,
  3. Lois K Lee2
  1. 1Division of Emergency Medicine and Transport, Children's Hospital Los Angeles, Los Angeles, California, USA
  2. 2Division of Emergency Medicine, Boston Children's Hospital, Boston, Massachusetts, USA
  1. Correspondence to Dr Saranya Srinivasan, Division of Emergency Medicine and Transport, Children's Hospital Los Angeles, 4650 Sunset Blvd, Mailstop #113, Los Angeles, CA 90027, USA; sasrinivasan{at}


Objectives Firearm injuries to children and adolescents remain an important cause of morbidity and mortality in the USA. The objectives of this study were to describe the prevalence of and epidemiologic risk factors associated with firearm injuries to children and adolescents evaluated in a nationally representative sample of US emergency departments and ambulatory care centres.

Study design We performed a retrospective cross-sectional analysis of data from the National Hospital Ambulatory Medical Care Survey from 2001 to 2010. Firearm injury-related visits in patients 0–19 years old were identified using the International Classification of Diseases, Ninth Revision, Clinical Modification firearm injury codes. The primary outcome was the prevalence of firearm-related injuries. We used multivariate logistic regression to analyse demographic risk factors associated with these injuries.

Results From 2001 to 2010, there were a total of 322 730 927 (95% CI 287 462 091 to 357 999 763) paediatric US outpatient visits; 198 969 visits (0.06%, 95% CI 120 727 to 277 211) were for firearm injuries. Fatal firearm injuries accounted for 2% of these visits; 36% were intentionally inflicted. There were increased odds of firearm injuries to men (OR 10.2, 95% CI 5.1 to 20.5), black children and adolescents (0–19 years) (OR 3.2, 95% CI 1.5 to 6.7) and adolescents 12–19 years old (all races) (OR 16.6, 95% CI 6.3 to 44.3) on multivariable analysis.

Conclusions Firearm injuries continue to be a substantial problem for US children and adolescents, with non-fatal rates 24% higher than previously reported. Increased odds for firearm-related visits were found in men, black children and those 12–19 years old.

  • Accident & Emergency
  • Epidemiology
  • Injury Prevention

Statistics from

What is already known on this topic

  • Firearm injuries are an important cause of morbidity and mortality in the paediatric population.

  • Adolescents, men and non-white children are at higher risk for firearm-related injuries.

What this study adds

  • This study demonstrates that the incidence of firearm-related injuries is 24% higher than previously reported estimates have suggested.

  • Of these injuries, 36% were intentional (homicide or suicide), while 64% were unintentional.


Firearm injuries to children and adolescents continue to be an important cause of morbidity and mortality in the USA. 1–3 From 1999 through 2007, firearm injuries accounted for 69% of homicides in children aged 1–19 years.4 Among 15–19-year-olds, 85% of homicides and 47% of suicides during this period involved a firearm.4 Prior studies have estimated that firearms are responsible for 4.9 fatal injuries and 19 non-fatal injuries per 100 000 children younger than 14 years old.1 ,4 ,5 For adolescents 15–19 years old, rates as high as 106.5 injuries per 100 000 have been reported.5 Some of the populations at increased risk for firearm injuries include adolescents aged 15–19 years, men, individuals of Hispanic ethnicity, those of black race and individuals with a history of involvement with the juvenile justice system.1 ,5–7

Costs and morbidity from firearm-related injuries remain substantial. The medical costs of treating firearm injuries exceed estimates of $2.3 billion dollars annually in the USA.8 In addition to the acute effects of injury, paediatric survivors of firearm-related injuries often sustain debilitating lifelong injuries, including permanent spinal cord and complex skeletal injuries, resulting in decreased mobility and difficulties with activities of daily living.1 ,7 ,9 Approximately 50% of paediatric patients hospitalised for a firearm-related injury are discharged home with a disability.7

Although previous studies have examined epidemiologic risk factors using local databases, many have primarily studied specific population groups thought to be at risk for firearm injuries and have not included nationally representative data.6 ,10–13 This study uses a database that provides firearm injury data using a nationally representative sample of US emergency department (ED) and ambulatory care centre visits, thus differentiating it from prior studies. The objectives of this study are to determine the national prevalence of firearm-related ED and ambulatory care centre visits and the epidemiologic risk factors associated with fatal and non-fatal firearm injuries in US children and adolescents less than 19 years old.


Study design and setting

We conducted a retrospective, cross-sectional analysis to identify firearm injuries in US children and adolescents by examining ED and ambulatory care visits in the National Hospital Ambulatory Medical Care Survey (NHAMCS) for the years 2001 through 2010. The NHAMCS is an annual survey conducted by the National Center for Health Statistics, a division of the Centers for Disease Control and Prevention (CDC). The survey is administered by the US Census Bureau and is comprised of data from a nationally representative sample of ED, outpatient, hospital-based and free-standing ambulatory surgery centre visits, excluding federal, military and Veterans Administration hospitals. Approximately 600 facilities are included in the survey each year and data are collected on more than 25 000 visits. Trained staff collects the data using standardised patient record forms during randomly assigned 4-week reporting periods. The NHAMCS database is available to the public via the internet.14

The NHAMCS survey uses a multistage probability design to generate national estimates by taking into account the probability of selection at each level (ie, the probability of selection of a given probability sampling unit (PSU)), the probability of selection of a hospital within the PSU, the probability of selection of an ED or clinic within the hospital and the probability of selection of a given patient visit within the ED or outpatient clinic).14 This study received institutional review board approval. This study did not receive any specific funding.

Study population

We identified patient visits for firearm-related injuries in children and adolescents 0–19 years old by the diagnosis and intent of firearm injury identified by the International Classification of Diseases, Ninth Revision, Clinical Modification codes (ICD-9-CM). The ICD-9-CM codes used included: undetermined injury by firearms (E985.0–E985.4), unintentional injury by firearms (E922.0–E922.3, E922.8–E922.9), intentional injury by firearms (E965.0–E965.4), suicide or self-inflicted injury by firearms (E955.0–E955.4, E 955.9) and injury due to legal intervention by firearms (E970). We excluded injuries caused by non-powder firearms (eg, paintball guns and air guns).


We collected patient demographic data on gender, race (white, black and other), ethnicity (Hispanic vs non-Hispanic), age and insurance type. The clinical variables included in the study were chief complaint, injury intent (intentional vs unintentional), outcome (fatal vs non-fatal) and disposition (discharged vs admitted). In addition, we collected facility-level data on the US region and metropolitan statistical area of the hospital. The NHAMCS divides the USA into four geographic regions (Midwest, Northeast, South and West) with comparable populations, which were included in this analysis. While NHAMCS does not include a direct measure of socioeconomic status, insurance type served as a surrogate measure. ‘Non-private’ insurance included individuals with Medicaid, Medicare, Workman's Compensation and self-pay. ‘Other’ insurance included those who were uninsured.14

Data analysis

Rates of firearm injuries per 100 000 individuals were calculated using population estimates from the US Census Bureau for gender, ethnicity, race and region from 2010.15 Descriptive frequencies were calculated. Univariate χ2 analysis was performed to compare the characteristics of firearm-related ED visits with those of non-firearm-related ED visits. For our multivariate logistic regression model, we decided a priori to include age, race, gender and region as well as variables with p<0.05 on univariate screen. Data analysis was performed using Stata V.10.1 (StataCorp, College Station, Texas, USA).


During the study period from 2001 through 2010, there were an estimated 322 730 927 (95% CI 287 462 091 to 357 999 763) ED and ambulatory care visits for children and adolescents 0–19 years old. Firearm-related injuries accounted for 0.06% or 198 969 (95% CI 120 727 to 277 211) visits. The average incidence was 19 897 firearm injuries per year with a rate of 23.9 injuries per 100 000 children. Of these injuries, 36% were intentional (homicide or suicide), while 64% were unintentional. The fatality rate was 0.4 fatalities per 100 000 children per year.

Children younger than 12 years of age accounted for 10% of all firearm-related ED visits, while children 12–19 years old accounted for 90% of these visits. There was a 2% fatality rate for firearm injury-related ED visits versus a rate of 0.03% for non-firearm-related visits (p<0.001) (table 1). The average rate of non-fatal firearm injuries was 23.5 injuries per 100 000 children per year.

Table 1

Demographics of paediatric firearm-related emergency department visits compared with all other paediatric ED visits

Children with firearm injuries in the ED were more likely to be men (p<0.0001), black (p<0.002), 12–19 years of age (p<0.0001) and uninsured (p<0.01) compared with children evaluated for non-firearm-related complaints or injuries (table 1). Rates of firearm injuries per 100 000 children stratified by demographic characteristics are presented in table 2. In the multivariate logistic regression model, male gender, black race and age 12 years and older were associated with increased odds of firearm injury-related ED visits (table 3).

Table 2

Rates of firearm-related injuries per 100 000 children

Table 3

Multivariate logistic regression analysis of firearm injury-related emergency department visits in children and adolescents


Our study provides an estimate of the incidence and prevalence of fatal and non-fatal firearm injuries among US children and adolescents aged 0–19 years presenting to an ED or ambulatory care site from 2001 to 2010. A total of 198 969 firearm-related injuries occurred during the 10 year study period for an average annual incidence of 19 897 firearm injuries. The rate of 23.5 non-fatal firearm-related injuries per 100 000 children is 24% higher than previous estimates of 19 non-fatal injuries per 100 000 children, indicating that the true scope of this problem may be larger than previous estimates would suggest.4 Nearly two-thirds of these injuries were unintentional. Men, children of black race and adolescents 12–19 years old had increased odds of firearm-related ED visits on multivariable analysis.

One possible explanation for our higher rate of firearm injuries is the utilisation of data from the NHAMCS, a nationally representative sample taken annually from 600 US ED and ambulatory care centres. While prior studies have reported rates of firearm injuries, many of these studies used local databases or registries and did not provide national estimates.6 ,9–12 Previous national studies have used data from the National Electronic Injury Surveillance System (NEISS) and Web-based Injury Statistics Query and Reporting System (WISQARS). NEISS is an annual probability-based sample of 100 US EDs from the Consumer Product Safety Commission.16 WISQARS is an interactive database run by the CDC and derives its data from NEISS.4 While both NHAMCS and NEISS provide national estimates of firearm injuries, NHAMCS provides a broader geographic sampling of hospitals and may capture data from regions that may not be included in NEISS.

Previously reported fatality rates have been in the range of 9–20%, while we report a fatality rate of 2%.1 ,11 ,12 The reason for this discrepancy again may be related to differences in data samples from which the estimates are derived. NHAMCS only includes deaths that occur after arrival to an ED or ambulatory care centre, but does not include deaths that occur outside the hospital or deaths of hospitalised patients not admitted through the ED. Prior studies have included fatalities from the National Vital Statistics System and other fatality databases, while our study did not, which may account for the lower fatality rate in our analysis.

Our study also demonstrated a lower hospitalisation rate than previous studies.5 ,7 ,17 This may be related to the bimodal nature of the severity of firearm injuries as children with firearm injuries often sustain either serious, life-threatening injuries resulting in death, or they may have less severe injuries and can be discharged home after ED treatment. Although we found a lower fatality rate than previous studies, it is important to note that ED visits for firearm injuries carry a fatality rate 67 times higher than the fatality rate for all other ED and ambulatory care visits.

Similar to previous studies, we also found that men, black children and adolescents aged 12–19 years had increased odds of sustaining a firearm-related injury.3 An increased risk of unintentional death, suicide and homicide among adolescents due to firearms has been demonstrated in regions where guns are more prevalent and in households that own firearms.18 ,19 Of the firearm injuries that occurred in our study, 90% were in children aged 12 years or older; therefore, injury prevention efforts should continue to be focused upon this group.

Prior studies have also demonstrated an increased risk of firearm injuries in children with a history of involvement with a juvenile justice system and in communities with lower socioeconomic statuses.1 ,3 ,5 ,7 ,11 ,20 Using insurance status as a proxy for socioeconomic status, we found an increased risk of firearm injuries in uninsured children on univariate, but not on multivariate analysis.

Of the injuries in this study, 64% were unintentional, which is higher than previously reported rates of unintentional injury ranging from 37% to 61.4%.1 ,5 ,12 ,13 ,21 Some of these unintentional shootings may occur when children are unsupervised in a home, find a loaded gun and accidentally fire it.22 An observational study of gun behaviour in 8–12-year-old boys found that 76% of the boys handled a gun after discovering one hidden in a drawer and that 48% of the boys pulled the trigger.22 Proper firearm storage practices, including keeping a gun locked and unloaded and storing ammunition locked and in a separate location from the firearm, are associated with a protective effect both for youth suicide and unintentional firearm injury.23 To this end, many states have enacted child access prevention laws, which impose criminal liability on adults who negligently store firearms, and these laws have been found to decrease firearm injury rates.24–27 With nearly two-thirds of the firearm injuries in our study being unintentional, advising caregivers about proper firearm storage is critical for reducing these unintentional firearm injuries.

Although proper firearm storage practices can help prevent unintentional firearm injuries, the American Academy of Pediatrics (AAP) affirms that the most effective approach for preventing paediatric firearm-related injuries is to advocate for the absence of guns from homes and communities with a legislative approach that serves to reduce the availability of firearms by imposing more stringent requirements upon gun owners.3 ,28 ,29 These laws, many of which have been enacted internationally, including in New Zealand, Australia, Austria, Brazil and Canada, have demonstrated reductions in rates of both youth suicide and homicide by firearms.26 ,30–35 An overall decrease in the availability and ownership of firearms is also associated with a decreased overall rate of suicide.36 Until the availability of firearms and access to firearms are reduced in the USA, the AAP advocates that guns be regulated as a consumer product in regard to child access, safety and design. Advances in gun engineering and design may also help decrease unintentional firearm injuries until regulations are enacted that further limit child access to firearms.3

Although some firearm regulations can serve to reduce firearm injuries, other forms of legislation may possibly increase firearm injuries in the paediatric population. Preliminary data suggest that states with ‘Stand Your Ground’ laws, which allow individuals to defend themselves using deadly force if they believe someone is trying to cause them serious harm, have increased rates of unintentional injury.37 Meanwhile, the ‘Privacy of Firearm Owners’ law, which forbids healthcare providers from questioning families about firearm ownership and storage practices, ignited a national debate in the USA about the First Amendment rights of physicians and their ability and responsibility to provide patients with preventative healthcare. This law was passed in the state of Florida, but is now under a permanent injunction. Despite the controversy caused by this law, several studies have shown that most parents are willing to discuss gun safety and that counselling by healthcare providers results in safer storage practices within the home.38 ,39 Therefore, paediatricians should continue to discuss firearm safety as part of routine healthcare maintenance visits and to advocate for legislation promoting safe firearm storage practices.25 ,28 ,29 ,40


Our study has several limitations. This is a retrospective analysis of NHAMCS, a national database that uses a probability-based sampling scheme to derive national estimates. As a result, it is possible that these derived estimates may not accurately reflect the true national values. In addition, limited clinical data are available from NHAMCS, which may result in misclassification bias due to limitations in coding. These factors make it more difficult to determine the true rate of intentional versus unintentional injuries. Finally, although 10 years of data were analysed, the numbers of observations in the subgroups were small, which made the estimates too unstable for further analyses stratifying by age group, injury intent and region.


This study demonstrates that firearm injuries to children and adolescents remain a significant problem for the US paediatric population with annual rates 24% higher than those that have been previously reported. To reduce firearm injuries to children and adolescents, more stringent firearm ownership regulations may be required, as has been demonstrated internationally. Increasing awareness of firearm injuries to children and adolescents, intensifying prevention efforts directed at safe firearm storage practices within the home and focusing on those at highest risk for these injuries are potential strategies to attempt to decrease firearm-related injuries to children and adolescents in the USA.


View Abstract


  • This work was presented at the American Academy of Pediatrics Annual Meeting, October 2011, Boston, Massachusetts.

  • Contributors LKL, RM and SS jointly conceived and designed this study. RM and SS jointly performed the data analysis. LKL, RM and SS interpreted the data. SS drafted the article. LKL and RM critically revised the manuscript and edited the article for intellectual context. All of the authors have seen and approved this version of the manuscript and take full responsibility for the manuscript as a whole.

  • Competing interests None.

  • Ethics approval Children's Hospital Boston institutional review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Atoms
    R Mark Beattie