Firearms in US homes as a risk factor for unintentional gunshot fatality

https://doi.org/10.1016/S0001-4575(02)00049-0Get rights and content

Abstract

This study used national data and a matched case-control design to estimate the relative risk of death by an unintentional gunshot associated with having firearms in the home. A sample of adults who died in the United States in 1993 from unintentional gunshot injuries was drawn from the National Mortality Followback Survey (NMFS) (n=84). Twenty controls were sought for each case from the 1994 National Health Interview Survey (NHIS) and matched to the cases by sex, age group, race, and region of residence (n=1451). Subjects were classified as having or not having guns in the home based interview responses. The relative risk of death by an unintentional gunshot injury, comparing subjects living in homes with and without guns, was 3.7 (95% confidence interval (CI)=1.9–7.2). Adjustment for covariates resulted in little change in the effect estimates. There was evidence of a dose-response effect: compared to subjects living in homes with no guns, the relative risk was 3.4 (95% CI=1.5–7.6) among subjects with one gun and 3.9 (95% CI=2.0–7.8) among subjects with multiple guns in the home. Having handguns in the home was associated with the largest effect estimates. Tests of homogeneity showed that the effect estimates did not vary significantly across categories of the matching variables. Firearms in the home appear to be a risk factor for unintentional gunshot fatality among adults. The magnitude of the observed effect estimates should be compared with those from additional studies.

Introduction

Firearms are present in approximately one-third of US households (Davis and Smith, 2000). A national estimate suggests that the privately owned firearm stock sums to nearly 200 million guns (Cook and Ludwig, 1997). Commonly cited reasons for firearm ownership are sport shooting, hunting, and self protection (Cook and Ludwig, 1997). Unfortunately, the handling and use of firearms can have serious unintended consequences that include unintentional gunshot injuries and death.

In 1989–1998, approximately 3500 children and 9300 adults died in unintentional shootings in the US (National Center for Health Statistics, 1989-1998). Although it has been declining annually, death from unintentional shootings remains an important public health problem. On average two people die each day in the US from unintentional gunshot injuries (Minino and Smith, 2001). Unintentional injuries that are fatal represent only part of the harm (Annest et al., 1995). For every unintentional gunshot fatality in the US, another 13 people are treated for non-fatal gunshot injuries in hospital emergency departments (Gotsch et al., 2001).

Evidence that firearm availability is a risk factor for homicide (Kellermann et al., 1993, Cummings et al., 1997) and suicide (e.g. Kellermann et al., 1992, Cummings et al., 1997) comes from multiple case-control studies. By contrast, case-control designs have not been used to investigate a link between firearm availability and death by unintentional shooting. State-level research provides evidence that such an association exists. In the US during 1979–1997, the rate of unintentional gunshot fatalities in states with the highest levels of gun availability was nine times higher than the rate in states with the lowest levels of gun availability (Miller et al., 2001). During 1988–1997, the rate of child unintentional gunshot deaths was 16 times higher in states where guns were most prevalent compared to states where guns were least prevalent (Miller et al., 2002). Results from a study of child deaths in California provide evidence that the association may stem from a gun kept in the decedent’s home. During 1977–1983, the prevalence of a handgun in homes of children killed in unintentional gunshot shootings was two times higher than predicted based on the level of handgun availability in homes in the region (Wintemute et al., 1989). This article presents a case-control study that further examined the topic.

Section snippets

Methods

The purpose was to derive an estimate of the relative risk for unintentional gunshot fatality associated with living in a home where firearms were or were not present. The subjects consisted of a sample of adults who died from unintentional gunshot injuries in the US and living controls. The cases came from the 1993 National Mortality Followback Survey (NMFS, 1993) (National Center for Health Statistics, 1998). The NMFS data combine information from surveys and death certificates and contain

Results

Eighty-four unintentional shooting fatality case subjects were identified. Twelve were killed with a handgun, four with a shotgun, three with a hunting rifle, and 65 with an unspecified type of gun. Twenty controls were successfully matched by sex, age group, race, and region to 63 of the 84 cases. The remaining cases were matched with between four and 18 controls (average=13). Table 1 shows characteristics of the cases and controls.

The prevalence of one or multiple firearms was considerably

Discussion

This study of a sample of adults in the US provides evidence that the presence of a gun in the home was a risk factor for unintentional shooting death. Having multiple guns appeared to compound the hazard, as did having handguns in particular. This study is the first on the topic, to my knowledge, to use a case-control design to estimate the magnitude of these relations. Hence, it will be important to compare the observed effect estimates with those derived from subsequent studies. Nonetheless,

Acknowledgements

I wish to acknowledge Susan B. Sorenson for conceptual advice, Hal Morgenstern for methodological advice, Colin Loftin for suggesting the NHIS as a source of controls, and Catherine Taylor, Katherine Vittes and anonymous reviewers for helpful suggestions, and Michael Mitcheu and the UCLA Academic Technology Services consultants. The methodological framework was developed in conjunction with my dissertation committee members: Joan Petersilia, Richard McCleary, Colin Loftin, Valerie Jenness, and

References (37)

  • G.L. Carter

    Accidental firearm fatalities and injuries among recreational hunters

    An. Emer. Med.

    (1989)
  • M. Miller et al.

    Firearm availability and unintentional firearm deaths

    Accident Anal. Prev.

    (2001)
  • J.L. Annest et al.

    National estimates of non-fatal firearm-related injuries: Beyond the tip of the iceberg

    J. Am. Med. Assoc.

    (1995)
  • Azrael, D., Miller, M., and Hemenway, D., 2000. Are household firearms stored safely? It depends on whom you ask....
  • Centers for Disease Control and Prevention. Sample Design of the 1993 National Mortality Followback Survey. No date....
  • T.B. Cole et al.

    Hunting firearm injuries, North Carolina

    Am. J. Public Health

    (1988)
  • Cook, P.J., Ludwig, J., 1997. Guns in America: National Survey on Private Ownership and Use of Firearms, NCJ 165476, US...
  • P. Cummings et al.

    The association between the purchase of a handgun and homicide or suicide

    Am. J. Public Health

    (1997)
  • Davis, J.A., Smith, T.W., 2000. General Social Surveys, 1972–2000. (machine-readable data file). Principal...
  • S. Frattaroli et al.

    Unintentional gun injuries, firearm design, and prevention: what we know, what we need to know, and what can be done

    J. Urban Health

    (2002)
  • Gotsch, K.E., Annest, J.L., Mercy, J.A., Ryan, G.W., 2001. Surveillance for fatal and non-fatal firearm-related...
  • Honaker, J., Joseph, A., King, G., Scheve, K., Singh, N., 2001. Amelia: A Program For Missing Data (windows version)....
  • L.D. Robinson et al.

    Some surprising results about covariate adjustment in logistic regression

    Int. Stat. Rev.

    (1991)
  • A.L. Kellermann et al.

    Validating survey responses to questions about gun ownership among owners of registered handguns

    Am. J. Epid.

    (1990)
  • A.L. Kellermann et al.

    Suicide in the home in relation to gun ownership

    New Engl. J. Med.

    (1992)
  • A.L. Kellermann et al.

    Gun ownership as a risk factor for homicide in the home

    New Engl. J. Med.

    (1993)
  • G. King et al.

    Estimating risk and rate levels, ratios and differences in case-control studies

    Stat. Med.

    (2002)
  • G. King et al.

    Analyzing incomplete political science data: An alternative algorithm for multiple imputation

    Am. Pol. Sci. Rev.

    (2001)
  • Cited by (49)

    • Firearm ownership, attitudes, and safe storage practices among a nationally representative sample of older U.S. adults age 50 to 80

      2022, Preventive Medicine
      Citation Excerpt :

      Ecologic studies demonstrate an association between firearm availability and adult firearm suicide and homicide rates (Miller et al., 2002a; Miller et al., 2002b; Dahlberg et al., 2004; Kellermann et al., 1992; Wiebe, 2003a; Hemenway and Miller, 2002; Miller et al., 2007; Sloan et al., 1988; Markush and Bartolucci, 1984; Marzuk et al., 1992), as well as with unintentional firearm fatality rates across the age spectrum (Miller et al., 2002c). Case control studies provide individual-level evidence that firearm ownership increases homicide and suicide risk (Kellermann et al., 1992; Kellermann et al., 1993; Bailey et al., 1997; Cummings et al., 1997a; Conwell et al., 2002), including for older adults (Conwell et al., 2002), suicide risk among adolescents (Brent et al., 1988), and unintentional firearm deaths in children, adolescents, and adults, including older adults (Wiebe, 2003b). These data also demonstrate that these risks extend beyond firearm owners to other household members, including children living in or visiting homes with firearms (Grossman et al., 1999; Cummings et al., 1997b).

    View all citing articles on Scopus
    View full text