Research Summary
Firearms have been the subject of longstanding controversies in American society, culture and law. There are many firearms in the United States — more than 300 million (Hepburn, Miller, Azrael, & Hemenway, 2007) — about as many guns as people. The United States has the highest rate of civilian firearm ownership in the world (Small Arms Survey, 2007). Most firearm owners own multiple firearms, with perhaps as few as 4% of the population owning 65% of the guns (Hepburn et al., 2007). Thus, in one recent survey, a minority of households (37%) included gun owners, with 24% of respondents reporting that they owned a gun and 13% reporting that another member of their household owned a gun (DeSilver, 2013).
Firearm violence takes a number of different forms, including, but not limited to, suicide and suicide attempts, violent conflicts and disputes, intimate partner violence, unintentional deaths and injuries, violent criminal activity, and violent acts while intensely distressed, intoxicated, or acutely psychotic. Firearms contribute significantly to homicide and suicide as causes of death in the United States, causing 11,078 homicides and 19,392 suicides in 2010, 11,101 homicides and 19,766 suicides in 2011 (Hoyert & Xu, 2012) . Mass shootings receive intense media coverage, generate understandable public alarm, and appear to fuel the purchase of firearms.1 However, the Congressional Research Service estimates that over the last 30 years (1983-2013) public mass shootings took 547 lives and left 476 victims injured, concluding that “while tragic and shocking, public mass shootings account for few of the murders or non-negligent homicides related to firearms that occur annually in the United States” (Bjelopera, Bagalman, Caldwell, Finklea, & McCallion, 2013, Summary section, para. 5). Thus, in order to be effective in reducing firearm violence, the national response must comprehensively address the phenomenon in its many forms.
The public health burden arising from firearms: Deaths, injuries, and other associated harms
Firearms are inherently dangerous and pose a substantial risk to the health of the public. Preliminary data for 2011, the most recent available, document that daily deaths from firearms averaged 54 suicides, 30 homicides, and more than 2 unintentional deaths, with totals of firearm-related deaths for the year equaling 19,766 of 38,285 suicides (52%), 11,101 of 15,953 homicides (70%), and 851 of 122,277 unintentional deaths (1%) (Hoyert & Xu, 2012). Additionally, for every firearm fatality, an estimated 2.25 persons suffer non-fatal injuries requiring emergency medical care or hospitalization (Gotsch, Annest, Mercy & Ryan, 2001). In 2012, there were 80,525 non-fatal firearm injuries: 4,068 self-inflicted injuries, 59,077 injuries from firearm assaults, and 17,362 unintentionally inflicted injuries (U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2014). The number of unintentional injuries and deaths may be underreported, especially among children (Luo & McIntire, 2013). One analysis estimated that gun violence imposed total costs of $174 billion on the United States in 2010, an average of $645 per gun in the United States, $5.1 million for each fatality, $433,000 for each gun injury requiring hospital admission, and $116,372 for each firearm injury requiring emergency department admission only (Miller, 2012). These estimates do not include the impact on those who endure consequences from witnessing or fearing firearm violence in their homes or communities when firearms are used to intimidate and coerce (Sorenson & Wiebe, 2004; Truman, 2011).
In a survey conducted in February 2013, 48% of firearm owners reported that they own guns for protection. This reflects a substantial change since 1999, when only 26% of gun owners reported that they own guns for protection and 49% of gun owners identified hunting/sport shooting as the primary reason they have a gun (Pew Research Center, 2013). Paradoxically, firearms owners have increasingly identified protection as their reason for acquiring a firearm even as rates of violent crime have dropped substantially. Violent crime rates have dropped by half since 1993 (U.S. Department of Justice, Federal Bureau of Investigation, 2013) and the 2012 murder rate of 4.7 per 100,000 persons compares to rates of 10.2 in 1980, 9.8 in 1991 and represents a decline of almost 17% since 2003 (U.S. Department of Justice, Federal Bureau of Investigation, 2013).
In addition, firearms are associated with increased risk. Purchase of a handgun is strongly associated with increased risk of suicide (Wintemute, Parham, Beaumont, Wright & Drake, 1999). Having a firearm in the home increases the likelihood of homicide or suicide of a family member (Dahlberg, Ikeda & Kreznow, 2004; Kellermann, et al. 1992; Kellermann, et al. 1998), including fatal shootings of women associated with intimate partner violence (Campbell, et al. 2003). Compared to other high-income countries, the United States by a substantial margin has the highest rates of firearm-related homicide, suicide and unintentional death and unintended injury among children and adolescents, leading the American Academy of Pediatrics to conclude that the “absence of guns from children's homes and communities is the most reliable and effective measure to prevent firearm-related injuries in children and adolescents” (American Academy of Pediatrics, 2012, e14160).
A public health approach to preventing deaths and injuries from firearms
A public health approach to the prevention of public health problems is a scientific approach. Scientists define a problem, conduct research to identify risk and protective factors, and use the knowledge about risk and protective factors to develop preventive interventions. The interventions are implemented and evaluated for effectiveness. The evaluation results then guide efforts to ensure the widespread adoption of effective programs and policies to mitigate risks or support protective factors. Public health approaches commonly utilize multidisciplinary collaborations among a range of stakeholders to identify and achieve goals for community and individual health and safety. Sound science generally does not provide definitive answers in one study or at a single point in time. Instead, scientific knowledge develops over time as new research clarifies and expands upon past understandings. Accordingly, in applying a science-based approach, one begins with the best available evidence and subjects it to ongoing, systematic scientific scrutiny.
The American Psychological Association published a report by an expert panel in December 2013 (American Psychological Association, 2013) in an effort to inform the public regarding the current science on firearm violence and contribute to national efforts to prevent firearm-related death and injury. The report reviews research on development, gender, and culture as antecedents to gun violence and what works to prevent gun violence at the individual, family, community, and societal levels.
Because access to a firearm is the common denominator in firearm violence, reducing access to firearms has been an important focus in prevention. Child access prevention laws , which hold adults criminally liable for unsafe storage of firearms around children, have reduced adolescent suicides and unintentional shootings of children (Webster & Starnes, 2000; Webster, Vernick, Zeoli, & Manganello, 2004). Undercover operations and lawsuits against dealers have reduced the diversion of firearms to criminals ( Webster, Bulzacchelli, Zeoli, & Vernick, 2006; Webster & Vernick, 2013). Other prevention efforts focused on access to firearms include design and manufacture of firearms such as “smart guns” that can be fired only by an authorized user, limitations on access to certain firearms such as assault rifles or products such as high-capacity magazines, and systems of distribution and sales that help prevent illegal diversion of firearms and “straw purchases” of firearms (Sorenson & Webster, 2013).
Access strategies also include regulating access to firearms for particular classes of persons. For example, U. S. law prohibits firearm purchase and possession by, among others, felons and persons dishonorably discharged from the military, subject to a domestic violence restraining order, or “adjudicated as a mental defective” or “committed to a mental institution” (Firearms, 2014) . These access restrictions are implemented by requiring federally licensed firearm dealers to request background checks on potential purchasers from the National Instant Criminal Background Check System (U.S. Department of Justice, Bureau of Alcohol, Tobacco, Firearms, and Explosives, 2005). One study has found that, if properly implemented, such prohibitions can significantly reduce violent offending among persons with histories of involuntary psychiatric commitment (Swanson, et al. 2013), but multiple impediments to their implementation have hampered their potential contribution to reducing firearm violence ( Kinscherff, Evans, Randazzo, & Cornell, 2013)
Some, but not all, educational interventions to reduce firearm violence have shown promise. Promising interventions include counseling by health care providers (especially when combined with distribution of cable locks to secure firearms) (Barkin, et al. 2008), police training for de-escalation of persons in crisis in high-risk situations (Teller, Munetz, Gil, & Ritter, 2006), and community, family, and individual interventions to promote healthy social development and reduce aggressive behavior among children and adolescents ( Cornell & Guerra, 2013) . On the other hand, e fforts to educate children about guns (largely to stay away from them), when tested with field experiments, indicate they are generally ineffective (e.g., Hardy, 2002). A “substantial body of scientific evidence [that] identifies important developmental, familial, and social risk factors for violence” ( Cornell & Guerra , p. 5) can guide the development of additional interventions.
The research on firearm violence indicates that while empirically-derived structured clinical judgment and actuarial tools have been shown to distinguish relative violence risk among researched populations (e.g., male domestic violence offenders, offenders with violence histories and mental disorders), no methods currently exist for reliably predicting whether or not specific individuals will behave violently, nor the specific time, place or manner (including firearm use) in which they will behave violently (Lidz, Mulvey, & Gardner, 1993; Meehl & Rosen, 1955; Monahan, et al. 2005; Nielssen, et al. 2009). On the other hand, science-based risk assessment and management strategies using empirically-derived assessment tools for individuals with histories of violence have developed as the standard for preventing targeted violence in many settings ( Kinscherff, Evans, Randazzo, & Cornell , 2013). In the behavioral threat assessment model, teams use highly individualized and situation-specific methods to prevent violence by specific persons identified as making or posing a threat of violence, including risk of using a firearm.
More research is required to guide policy and practice since some promising measures have not yet been shown to be effective or may have unintended consequences (National Research Council, 2005; Institute of Medicine and National Research Council, 2013). Research can also help determine which initiatives are ineffective in reducing harm from firearm violence. For example, “buy-back” programs might raise public awareness of gun violence, but have been ineffective in reducing firearm deaths (Institute of Medicine and National Research Council, 2013; Makarios & Pratt, 2012). Additionally, controversies persist as to whether various legal requirements for mental health professionals to “warn or protect” when providing services to potentially violent persons are more likely to reduce violence or deter persons from seeking mental health care (Kinscherff et al, 2013).
Some research suggests that more rigorous reporting and background checks of persons whose mental health history disqualifies them from firearms ownership lowers risk of violent criminal offending (Swanson, et al. 2013 ). Yet, concerns persist about the risks of stigmatizing persons with mental illness while also potentially fostering public perceptions that firearm violence can be readily reduced to a “mental illness” problem (e.g., Appelbaum, 2013). Research could also help determine the effects of recent legislative efforts to bar medical care providers from asking patients about firearm possession and access (Medical privacy concerning firearms; prohibitions; penalties; exceptions, 2014; The Patient Protection and Affordable Care Act, 2010 ). Program developers and sponsors are encouraged to articulate clear rationales for policies, programs, and practices and to evaluate them.
There are some important barriers to the scientific research needed for a comprehensive public approach to the prevention of firearm violence. First, a universal system for collecting data on incidents of firearm violence does not exist. Several Institute of Medicine and National Research Council reports have identified the National Violent Death Reporting System as a promising approach for gathering essential data on firearm violence (National Research Council, 2002; National Research Council, 2005; Institute of Medicine and National Research Council, 2013), yet this system currently includes data from only 16 states (U.S. Department of Health and Human Services, Centers for Disease Control and Prevention , 2013).
Second, state and federal restrictions, including restrictions on the Centers for Disease Control and Prevention (Omnibus Consolidated Appropriations Act, 1996) and the National Institutes of Health (Consolidated Appropriations Act, 2012), limit or discourage firearm violence research and preclude or discourage the collection and use of firearm violence information (Institute of Medicine and National Research Council, 2013; Kellermann & Rivara, 2013). These restrictions exist despite widely-accepted and widely-implemented research practices that safeguard the privacy of individuals (e.g., medical records research) in order to gather information on the occurrence of other public health problems (e.g., certain infectious diseases). If federal and state restrictions on such data are removed, research by psychologists and others can be used to devise, implement, and evaluate research-based public health approaches to firearms-related death and injury.
Firearm violence and diversity
As noted above, access to a firearm is the common denominator in every firearm-related death or injury. Beyond this obvious fact, achieving a greater understanding of the different forms of firearm violence, the populations disproportionately harmed, and the factors relevant to preventive interventions will involve addressing considerable complexity. Firearm violence disproportionately affects specific groups within the United States. Patterns of injury and death from firearms (attempted and completed homicides and suicides, and unintentional injuries) differ according to factors including age, gender, gender identity and expression, sexual orientation, race and ethnicity, geographic region and locality, educational level, employment status, job and working conditions, income level, and social class (Hepburn & Hemenway, 2004; Institute of Medicine and National Research Council, 2013; Jenkins, 1996; Kegler & Mercy, 2013; Kennedy, Kawachi, Prothrow-Stith, Lochner, & Gupta, 1998; Loomis, Marshall & Ta, 2005; Nock, Borges, Bromet, Cha, Kessler, & Lee, 2008; Peek-Asa, Erickson, & Kraus, 1999). These disparities reflect a complex interaction of multiple risk, protective , and contextual factors at individual, community , and societal levels, including differential access to resources that promote health and safety (Krug, Dahlberg, Mercy, Zwi, & Lozano-Ascencio, 2002).
For example, firearms are the most frequent means of suicide among older adult white men and contribute to them having a very high suicide rate (U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2013). White adolescent males also have elevated rates of suicide by firearms (U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2013). Young Black males living in impoverished urban communities bear the greatest risk of homicide by firearms (Hammond & Prothrow-Stith, 2001). Women are disproportionately more likely to be the victims of firearm violence by an intimate partner (Sorenson, 2006). In order to be developmentally and culturally appropriate and attentive to relevant aspects of diversity, public policy and prevention strategies must attend to the different relative risks, occurrence, and contexts across groups.
Firearm violence and mental illness
There is little research specifically on firearm violence among persons with severe 2 mental illness, but there is a relatively large literature on the relationship between severe mental illness and violence in general. As one commentator has put it, “[t]he vast majority of people with mental disorders do not engage in violence and the proportion of overall risk of violence attributable to mental disorders is small…The best U.S. data put the population attributable risk for violence due to mental disorder between 3% and 5%” (Appelbaum, 2013, p. 565, citing Swanson, 1994). Research has demonstrated a modest association of mental disorders with increased violence, with greatest risk for those who have additional risk factors such as substance abuse, histories of being violently victimized, continued exposures to violence, antisocial personality traits , and histories of involuntary commitment (Appelbaum, 2013; Swanson, et al. 2002 ; Swanson, et al. 2013). Some recent research points to substance abuse as the primary factor in violence risk among persons with severe mental illnesses (Fazel, Gulati, Linsell, Geddes, & Grann, 2009; Fazel, Langstrom, Hjern, Grann, & Lichtenstein, 2009). One recent large longitudinal study found no significant independent association between severe mental illness and subsequent violent behavior, rather that people with severe mental illness had a greater likelihood of having other risk factors associated with violence, for example, a history of violent victimization and substance abuse (Elbogen & Johnson, 2009). Research has found that access to adequate mental health treatment in the community following psychiatric hospitalization reduced subsequent violent acts, most of which were minor assaults unlikely to significantly injure and were directed at known persons rather than strangers (Monahan, et al. 2001). In contrast to the small association between mental illness and violence directed at others, there is a significantly elevated risk of suicide among persons with mental disorders with some 90% of persons who commit suicide having some combination of symptoms of depression, other mental disorder, and substance abuse (Moscicki, 2001).
Aggressive acts often are distinguished by whether they are planned to achieve anticipated goals (variously termed instrumental, targeted or predatory aggression) or whether they are driven by intense emotions at a time of crisis (variously termed affective, impulsive, reactive, or hostile aggression). Persons engaging in impulsive violence are sometimes described as acting “in the heat of the moment” or “without considering the consequences.” Predatory and impulsive aggression implicate different neurological systems with impulsive aggression characterized by high levels of autonomic arousal and negative emotions such as fear or anger, usually in response to a perceived provocation or stressor (Siever, 2008).
Whether or not they suffer from a mental disorder, persons in intense emotional crisis are at higher risk of impulsive aggression and harming themselves (Wyder & De Leo, 2007; Zouk, Tousignant, Seguin, Lesage, & Turecki, 2006) or others (Meloy, 2006), including spouses (Edwards, Scott, Yarvis, Paizis, & Panizzon, 2003) and children (Fujiwara, Barber, Schaechter, & Hemenway, 2009; Rodriguez & Richardson, 2007). Persons in crisis include those experiencing desperation, despair, panic, rage or other intense emotions that may result in acts of impulsive violence involving a firearm, if one is accessible. Some participants in contemporary policy debates about firearm violence seem to assume that “bad guys” cause violence and the appropriate way to deal with violence is to arm “good guys” to deal with “bad guys.” Unfortunately, this approach will not be effective in preventing violence, because “ an armed 'good guy‘ can become a 'bad guy‘ [who uses] a gun in a moment of temporary despondence or rage (Bandeira, 2013; Wintemute, 2013)” (Sorenson & Webster, 2013, p. 32).
One priority of the public sector mental health system is to meet the needs of people with severe mental illnesses. However, the public mental health system also has a second priority: to respond to serious emotional crises that can happen to anyone at any time, and especially to respond at times when these crises produce elevated risks of harm to self or others. Yet, the steady loss of $4 billion from state mental health budgets since 2008 (Appelbaum, 2013) has eroded the capacity to respond in a reliable, timely, and competent manner to those in intense emotional crisis. The broader system of mental health services beyond the public sector also has a crucial role to play in responding both to persons with severe mental illness and to persons in serious emotional crisis.
For the mental health system to play an optimal role in preventing firearm violence, policy makers will need to increase mental health care resources. However, in promoting such increases, policy-makers and advocates should take care to address both priorities of the mental health care system, to encourage help-seeking behavior by persons in crisis and by persons with mental illness, and to avoid reinforcing the stigma that both groups experience. Policymakers and advocates should go beyond the concern regarding mass shootings to focus upon the contribution of depression to the higher rates of suicide and the contribution of emotional crisis to incidents of gun violence.
The American Psychological Association Panel of Experts Report on Gun Violence (American Psychological Association, 2013) discusses various policies seeking to restrict access to firearms by persons with mental disorders. Many of these policies have been criticized for using broad criteria that bear little relationship to actual risk and for failing to identify persons who may pose significant or imminent threat of violence (Fisher & Lieberman, 2013 ). However, some research indicates such policies, if properly implemented, can significantly reduce violent offending among persons with histories of involuntary psychiatric commitment (Swanson, et al. 2013). Policy interventions such as these warrant ongoing evaluation given their potential for both reducing firearm violence and inadvertently deterring persons from seeking mental health care or being frank with their clinical care providers about risk factors for firearm violence.
1 On December 14, 2012, a gunman killed 20 first grade students and 6 school personnel at Sandy Hook Elementary School in Connecticut and wounded two others. In the next week, December 17-23, 2012, the most requests for background checks (a proxy for gun sales) in a week since 1998 were submitted to the National Instant Criminal Background Check System, 953,613 requests, nearly 50% larger than the next highest week (U.S. Department of Justice, Federal Bureau of Investigation, 2013).
2 The terms “serious mental illness” and “severe mental illness” are both commonly found in research literature. The term “severe mental illness” is used here for the sake of consistency.