Research Project
10400411
PROJECT SUMMARY/ABSTRACT In the United States, personal protection is the leading motivation for purchasing and owning firearms, especially handguns. However, available evidence mostly suggests that residents of homes with guns are more?not less?likely to die from homicide. A key limitation of the evidence base is that individual- level studies of the relationship between firearm access and lethal assaults have been relatively small in scale, permitting estimation only of average population-wide effects. Such estimates may miss protective effects (or disproportionately high risks) among certain subgroups?women, for example, or people living in dangerous neighborhoods. The goal of this study is to explore the relationship between living in a home with lawfully-owned handguns and the risk of dying by homicide?overall, and for a variety of distinct subgroups. These analyses are made feasible by a unique, recently-constructed cohort of 28.6 million adults residents of California who are followed for up to 12.2 years. The cohort consists of a mix of handgun owners, nonowners who live with owners, and nonowners who live in homes without handguns. We observe individual-level, time-varying information on cohort members? handgun purchasing history (back to 1985), cause-specific mortality, and demographic characteristics; household identifiers permit linkage of cohabitating cohort members. The study will focus on homicides occurring in and around the victim?s home. We will begin by comparing rates of homicide victimization in homes with and without handguns. Next, we will repeat this analysis separately for men and women; for people of different age groups and race and ethnic groups; for household members who own handguns (predominantly men), and those who don?t own but reside with owners (predominantly women); and for people living in relatively dangerous and relatively safe neighborhoods. We will also assess how risks of different types of homicide (e.g., homicides perpetrated by family members vs by strangers) vary between homes with and without handguns. All of these associations will be estimated by fitting extended Cox proportional hazards models that adjust for age, sex, race and ethnic group, and long-gun ownership; the models will allow baseline hazards to vary according to neighborhood. This is the first cohort study to investigate the relationship between household exposure to firearms and risks of homicide. Study results will provide policymakers, law enforcement, public health practitioners, and? most importantly?handgun owners and their cohabitants with a more complete and accurate accounting of the risks and benefits of living in a home with handgun. The findings will also inform firearm violence prevention initiatives: more specific information on who is at especially high risk of homicide can be used to target interventions and tailor public health education and messaging.
