Despite advances in the science of violence prevention, several gaps and challenges remain. The impact of violence is mediated through several means, such as firearms and pesticides, and modifiers, such as alcohol. Specifically, the lethality of firearms and commonly used pesticides result in higher fatalities, while alcohol reduces inhibitions that might otherwise be a barrier to violence. Speakers presented on how these means and modifiers affect violence and violence prevention, and how reducing access to these means can reduce violence.
Speakers discussed issues around the access to lethal means of violence and successful methods to reduce such access, particularly among those with mental illness. Speakers also noted, however, that such violence tends to be self-directed rather than other-directed. As noted by other speakers in the workshop, people with mental illness are not at higher risk of interpersonal violence compared with the general population, but their risk of self-directed violence is much higher. Thus, speakers discussed how restricting access to lethal means could be a method of reducing the incidence of suicide and self-directed harm.
Firearms Means Restriction and Mental Health1
Daniel Webster considered the research on firearms prohibitions on people with mental illness. A study conducted by Jeff Swanson in Connecticut analyzed data from public mental health and criminal justice agencies, providing a sample of 23,000 individuals with severe mental illness (Swanson et al., 2013). Criteria for severe mental illness included diagnosis and hospitalization for schizophrenia, bipolar disorder, and major depressive disorder. The study focused on two general cohorts: one had at least one of four potential mental illness disqualifications for firearms possession, the other had no disqualifications. Of the overall sample, only 5 percent were disqualified because of severe mental illness. In 2007, the state began recording these disqualifications. The study compared the rates of violence in both groups before and after the onset of reporting. Swanson and his colleagues found an odds ratio of 0.69; essentially, there was a 31 percent reduction in risk for arrest of violent crime because of the reporting, while those who were not affected by reporting had no change in offending (Swanson et al., 2013). Webster pointed out that while crime in this study involved more than just gun-related incidents, it still showed a significant impact on gun-related crime. Other studies looking at interpersonal violence and other mental health issues, including substance abuse, also show that mental-illness-related prohibitions reduce violence in those two groups. In particular, one study looked at perpetrators of intimate partner violence and found reductions in homicide associated with firearm restrictions for those with restraining orders (Vigdor and Mercy, 2006).
However, most of the risk related to mental health and violence is around self-harm and access to lethal means. Webster noted that there is a common perception that someone with intent of suicide will find a way to do so regardless of available methods. In contrast, several historical examples of restriction of lethal means and subsequent reduction of suicide. For example, the removal of coal ovens in British homes reduced suicides by one-third, and raising the barrier on the Duke Ellington Bridge in Washington, DC, reduced suicides by one-half. The success of a suicide attempt is also related to the lethality of the method. Despite the high availability of means such as rope, knives, and poison, these make up a lower percentage of the case fatality rate compared with firearms.
Bringing these two factors together suggests that suicide risk associated with access to lethal means is higher. In a national study, Doug Weibe found that, controlling for other factors, risk is elevated threefold when there is a gun in the home (Wiebe, 2003), a finding shown in other research as well.
1 This section summarizes information presented by Daniel Webster, Johns Hopkins Bloomberg School of Public Health.
Other studies, both cross-sectional and longitudinal, that look at the population level have found a positive association with prevalence of firearms ownership and risk of suicide, with higher risks at younger ages2 (Miller et al., 2007; Stevens et al., 2006). On a similar note, an analysis of gun ownership rates in the 1990s and youth suicide rates showed that suicides decreased dramatically as household gun rates dropped.
Regarding restricting firearms access to youth, the laws that require owners to lock guns away reduced suicide risk among 14- to 17-year-olds by 8 percent overall. Additional laws related to restricting firearms did not have an effect, though, nor did they have an effect on older youth aged 22 to 24 years3 (Webster et al., 2004). Other studies support these findings, showing protective effects for older individuals with the Brady Handgun Violence Prevention Act and some state laws that require permits for purchasing (Andrés and Hempstead, 2011).
Means Restriction and Suicide4
Suicide accounts for 60 percent of all violent deaths in men, and 75 percent in women, globally. In high-income countries, more than 80 percent of violent deaths are suicide. In sub-Saharan Africa and Latin America, suicide accounts for a lower proportion of violent deaths, but in Asia and Eastern Europe, suicide accounts for more than half of violent deaths. Of all suicides globally, 84 percent occur in low- and middle-income countries. China and India account for 56 percent of all male suicides and 61 percent of all female suicides. However, most of the research and intervention models on suicide prevention come from high-income countries.
In both China and India, 50 percent of suicides are due to poison, usually pesticides or rodent poisons, both of which are lethal. In a study that looked at suicides globally, researchers found that one-third used poison, usually found in the home (Gunnell et al., 2007). Phillips described a few characteristics of pesticide poisoning and suicide in China:
- Relative to other methods, it is lethal—6 percent of those who end up in the emergency department die, versus 1 percent of those who use other methods.
- 43 percent of those who use pesticides thought about the attempt for less than 5 minutes, compared with 16 percent of those who
4 This section summarizes information presented by Michael Phillips, Shanghai Jiao Tong University School of Medicine.
used other methods of suicide; this is partly because of the ready availability of such pesticides.
- Those who use pesticides tend to have low intent to die, which contradicts the popular opinion that those who are more intent on dying use more lethal methods.
- Only 33 percent of those who ingest a pesticide had a diagnosis of mental illness at the time, compared with 54 percent of those who use other methods.
Michael Phillips emphasized that last point by addressing “common knowledge” in the West that suicide is an outcome of mental illness. From their research, Phillips and his colleagues observed that more than 60 percent of those who attempt suicide do not in fact have mental illness, and using psychological autopsy, they found that 30 percent of those who died of suicide did not have a mental illness.
Regions in China where pesticides are stored in homes have higher rates of suicide than regions where pesticides are less likely to be stored at home. At the same time, as China’s population has increasingly urbanized, with fewer people working in agriculture, access to pesticides has dropped, and suicides have reduced. Phillips remarked that there are 100,000 fewer suicides in China per year than 20 years ago. He hypothesized that in low- and middle-income countries with a large agricultural sector, patterns of pesticide access and use might be a more useful approach to suicide reduction than prevalence and treatment rates of mental disorders. As such, he proposed a set of strategies to address suicide by pesticide poisoning:
- Banning the most toxic compounds
- Decreasing access to pesticides in the home
- Community education about the lethality of these chemicals and about appropriate storage
- Improved training and increased access to necessary drugs and equipment for rural primary care health providers
While improving medical knowledge and treatment and raising community awareness are important, Phillips asserted that means restriction should be the main focus of efforts to reduce suicide. The effectiveness of means restriction, however, is dependent on how feasible the restriction is, and what the proportion of deaths that particular method comprises. It also depends on whether a substitution method is available. As an example, Phillips described a study in Sri Lanka in which a more lethal chemical was substituted for one that was banned; yet, when the importation of highly toxic pesticides was completely banned, the suicide rate dropped
dramatically. He cautioned that a means-restriction approach should be constantly monitored for these and other types of mediating factors.
Phillips also suggested additional large-scale interventions for restricting means:
- Promote secure storage in homes, fields, or a centralized community location
- Establish a minimum pesticide list, so individuals may only own certain pesticides
- Promote integrated pest management programs
- Apply a tax to pesticides that increases with pesticide lethality
- Limit usage of pesticides in each village or community to a small number of licensed individuals who would apply pesticides for all community members
- Train pesticide retailers to recognize potentially suicidal individuals
- Limit sale of pesticides to single-use amounts
Phillips closed with a description of a project he and his colleagues implemented in Shaanxi province in China. The objective of the project was to promote the installation of 10,000 lockboxes for the storage of pesticides. An educational campaign was also rolled out and suicide rates before and after the intervention were monitored and compared with other townships without the intervention. The lockboxes had two keys, with the idea that two people (usually husband and wife) were required to open the box. Researchers followed the families over 3 years to assess compliance. They found that there was limited uptake of the intervention—about 20 to 30 percent. People were using the box, but few were using the locks. When the educational component stopped, use of the locks dropped: 88 percent were using the box, but almost none of them were locking it after the 3-year period. Ultimately, they did see a drop in suicide rates in the intervention areas—about 23 percent—while the rates in the control sites increased by 2 percent. Phillips observed that 100 percent compliance is unreasonable to expect, and means restriction needs to be part of a larger overall suicide prevention strategy. Yet, he noted that in low- and middle-income countries, focusing on individual-level mental health approaches might not be the best use of resources.
Regulation and Means Restriction5
Mike Luo described issues raised by several news stories he wrote regarding gun violence and mental illness. In particular, in examining mass
5 This section summarizes information presented by Mike Luo, The New York Times.
violence events, such as one in which a disturbed young man opened fire on a crowd, killing several people, he questioned what it meant for people with mental illness to have access to firearms. He noted that the current federal standard stipulates that one cannot purchase or possess firearms if one has been involuntarily committed or adjudicated as “mentally defective.” The vast majority of those with mental illness, even severe mental illness, will never get to this point. Luo and his colleagues wanted to explore this area further to learn the stories of people with diagnosed mental illness who possessed firearms. However, he noted, there was a big privacy challenge in this area, dealing with both mental health and gun ownership.
In most states, he observed, records of purchases of concealed handgun permits are not publicly available. Because he was unable to obtain such records via public inquiries, he instead inquired at police departments and courts for records of people from whom firearms were confiscated for mental health reasons. Such calls are not enough to disqualify someone from possessing firearms but are usually grounds for temporary confiscation because a person is a danger to himself or herself or to others. However, the circumstances under which it is legal to confiscate a firearm are not straightforward. Luo noted that while taking away a firearm on someone’s person is usually allowable, the situation is less clear when the firearm is in another location. Most police departments would require a warrant to confiscate the weapon in these circumstances; for example, Connecticut and Indiana passed laws giving police more leeway on this.
In the past year in Connecticut, there were 180 instances when police removed firearms from people they deemed to pose a risk of imminent danger, 40 percent of which involved serious mental illness. In 2012, Luo and his colleagues found that in Marion County, Indiana, there were 30 instances of confiscation, with about 40 percent of those involving mental illness. Most people were placed under observation, but not involuntarily committed, and in most instances, the firearms were returned shortly.
Luo cited a few examples of these policies at work. In Indianapolis, before the law giving police greater jurisdiction was passed, an individual with a diagnosis of schizophrenia retook possession of his firearm and was later involved in a police shooting. In Hillsborough County, Florida, there was another instance in which a veteran with a history of treatment for depression, anxiety, and paranoia made violence-related comments to his psychiatrist and subsequently had his firearms confiscated. He was involuntarily hospitalized but not committed, and a few months later had his firearms returned. In a third situation in Colorado, an individual with prior suicide attempts, who also had not been committed, had successfully requested his firearms be returned.
In the context of these cases, Luo raised a series of questions about the intersection of privacy, regulation, public health, science, and rights. Given
that involuntary commitment is a difficult process, where is the standard set for disqualification of possession of firearms? Should there be a protocol for restoring rights, and what should it look like? What is the best way to predict future violence by a person with mental illness and in possession of a gun? Where is the line between Second Amendment rights and public health and safety?
In the subsequent discussion, workshop participants shared their perspectives on issues raised during the presentations, particularly around predicting future violence and restricting lethal means. Webster noted that, rather than focusing on diagnosis or involuntary commitment to disqualify an individual from owning a lethal weapon, the focus should be on potential danger. For example, indicators of substance abuse, such as multiple violations for driving under the influence of alcohol (DUIs), magnify risk associated with severe mental illness. Mark Rosenberg of The Task Force for Global Health agreed, and he suggested that stress and distress are triggers for suicide and could be assessed.
Phillips echoed comments made earlier, stating that predicting individual behavior is very difficult, and current instruments to do so are imprecise. Luo concurred, mentioning that most assessments of risk are conducted by psychiatrists using unstructured criteria.
Eric Caine of the University of Rochester Medical Center considered whether community-based approaches, in which all members of the community feel invested, might be a more effective means of reducing violence. He spoke about a program in King County, Washington, in which a coalition was built among public health officials, injury prevention stakeholders, and firearms retailers. The program involves incentivizing firearms buyers to purchase gun locks by offering discounts. It does not ask anyone to serve as a gatekeeper, but instead builds a community of safety. Caine suggested that such an approach might help bridge differences among different stakeholders.
Alcohol’s impact on violence is observable but not fully understood. Speakers discussed insights from experimental and observational studies that explore the neurobiological and sociological pathways of alcohol-mediated violence. They also discussed gaps in policies, both in the United States and around the world, including those related to lack of funding and political and social will.
Escalated Aggression in Rodent Models:
Novel Brain Mechanisms for Alcohol6
Evidence suggests that alcohol plays a significant role in violence, but the relationship is complicated, Klaus Miczek explained. For example, two-thirds of all violence involves alcohol: 86 percent of homicides, 60 percent of sexual offenses, 75 percent of spousal abuse, and 30 percent of assault offenders report using alcohol.
In the 1980s, a watershed discovery illuminated the mechanism by which alcohol affects the brain, which caused a shift toward focusing on specific proteins, such as glutamate, GABA, and serotonin. He noted that alcohol has a biphasic dose-effect: at low and acute doses it has pro-aggressive effects, and at high doses it is anti-aggressive. Alcohol withdrawal also causes aggression; the greater the exposure to alcohol, the more intense the withdrawal and the more intense the aggressive episodes. However, the impact of alcohol on individuals is highly variable. In a certain subset of individuals, alcohol causes a large change in aggressive behavior. Miczek asked, “Who are these individuals, how can they be identified, and can they be corrected?”
In mouse studies, researchers have observed that some mice display large increases in aggressive behavior, as well as a change in the pattern of that behavior. In an animal that consumes water, the typical pattern for biting is in the rump. For those who ingest alcohol, the shift in target is dramatic: not only are they biting more frequently, but the bites themselves cause greater injury.
Miczek expounded further on the mechanisms by which these behavior shifts occur:
- Dopamine. Aggression produces dangers, but it also produces satisfaction and pleasure associated with rewards. In animal studies, it is possible to measure neurochemical events before, during, and after an aggressive act, as well as during recovery. In anticipation of an aggressive episode, dopamine rises in the nucleus accumbens while serotonin drops in the cortex. Researchers provoke an aggressive episode at a specific time of day for 10 days, and then do nothing on the 11th day. What they observe is the same neurochemical change in the rodent’s brain on the last day, despite the event not occurring. In effect, the rodent has been conditioned.
- Hypothalamus stimulation. Researchers discovered a locus of aggression in the hypothalamus by injecting a virus carrying a light-sensitive protein into the brain and then stimulating the
6 This section summarizes information presented by Klaus Miczek, Tufts University.
protein. Afterward, the rodents can be triggered, by flashing a light, to attack both animate and inanimate objects. Miczek suggested that these findings could be explored further in the alcohol animal research.
- Serotonin. Numerous studies over the years have focused on serotonin and its effects on aggression and violence. Findings have been inconsistent because serotonin is a complex molecule. Serotonin neurons in the brainstem project to other parts of the brain in a segregated manner, so they can be individually turned off by inserting toxins in specific places. Additionally, only serotonin neurons that originate in subregions that also express dopamine receptions are important in aggressive behavior, indicating the system consists of parallel processes. Focusing on the relevant serotonin process has implications for research on the impact of alcohol.
- GABA. The GABA (γ-aminobutyric acid) receptor is the target of action for alcohol. Alcohol acts as a positive allosteric modulator to facilitate the action of GABA, an inhibitory neurotransmitter. Because of the biphasic effect of alcohol (i.e., pro-aggressive at low doses, sedative at high doses), it was originally thought that different mechanisms were involved. In the 1990s, researchers discovered that different genes encode subunits of the GABAA receptor, with the α-2 subunits related to pro-aggressive effects.
- Glutamate. The N-methyl-D-aspartate (NMDA) receptor is one of several glutamate receptors in the brain. It has a number of subunits, which can be targeted to alter psychotic episodes. In particular, the use of a specific Alzheimer drug, memantine, enhances aggression in individuals who do not show heightened aggression, but produces no effect on those who do. Glutamate and GABA act as a go-and-stop mechanism for serotonin, including serotonin subsystems responsible for heightened aggression.
- Neuroendocrine factors. There is some promising evidence that corticotropin-releasing factor receptor 1 (CRF1) has a calming effect on mice with heightened aggression, possibly by mediating an aggression-related serotonin pathway.
Miczek closed by emphasizing the important role that the mouse model played in teasing apart the various pathways in the brain that result in aggressive behavior, particularly in relation to alcohol consumption, in a manner not possible in human research.
Alcohol Use and Intimate Partner Violence7
Kenneth Leonard remarked that the association between alcohol use and interpersonal violence has been observed in several cultures and contexts around the world, regardless of strictness of alcohol norms, level of violence, or other cultural overlays. This relationship has also been seen in multiple samples, including a nationally representative sample and criminal and clinical populations. It is an association of moderate strength and observed longitudinally, even when other factors related to aggression and conflict within relationships are controlled (Leonard and Senchak, 1996).
While there are effects of chronic use of alcohol, much of the aggression-related effects stem from the acute use of alcohol. Leonard noted that this can be studied in two ways: event-based research and experimental studies (largely in college students administered alcohol). Event-based research would include the examination of an event of partner violence, and a comparable event, such as a severe argument that did not result in violence. In a study Leonard conducted with his colleagues, couples were queried on alcohol consumption during conflicts involving verbal aggression, moderate physical aggression, and severe aggression. The husband’s drinking was strongly associated with severe violence, while the wife’s drinking was less clear (Leonard and Quigley, 1999). In another study, men and women described conflict episodes that did or did not involve violence in the months before alcohol abuse treatment. In the events that had physical conflict or violence, the husband’s use of alcohol was higher (Murphy et al., 2005).
In experimental studies of alcohol and aggression, the aggression that is provoked is usually mild and not necessarily similar to violence seen in the community. Many of them include competitions that involve the “reward” of inflicting mild harm (e.g., a mild shock) on their opponent. Alcohol consumption tends to result in a more intense shock, an effect that is dose-dependent, while the placebo (i.e., no alcohol) has no effect (Bushman, 1997; Ito et al., 1996).
Other types of experimental studies look at alcohol use and aggressive verbal behavior. In the context of intimate partner violence, Leonard remarked, this makes sense as violence often emerges out of an ongoing verbal conflict. In one study, couples were invited to discuss a previously agreed upon topic of conflict, to establish a baseline. They were then separated for a period of time, and the husband was given alcohol. When they returned to discuss another conflict, the interaction was marked by very high levels of negativity on the part of both the husband and the wife, an effect not seen when a placebo was administered (Leonard and Roberts,
7 This section summarizes information presented by Kenneth Leonard, Research Institute on Addictions.
Leonard proposed a cognitive disruption model to explain alcohol’s impact on aggression. Intoxication leads to some level of cognitive impairment, to which people adapt by focusing on salient cues and missing subtle context. This theory suggests that alcohol should exacerbate overt emotions in certain settings—that is, a setting that evokes aggression would be heightened with alcohol, whereas a situation that evokes sadness would have low risk of aggression. Alcohol exacerbates a person’s reaction to the most dominant cues and hides those that are peripheral and could be inhibiting (Parrott and Giancola, 2004). To the extent that mental illness is associated with negative affect and impaired self-control, Leonard postulated, alcohol might interact with psychopathology to create a high risk of violence. He showed one study that suggested a synergistic effect for substance use disorders and mental illness (Van Dorn et al., 2012).
Finally, successful treatment of alcohol use disorders results in a reduction of aggression. If sobriety levels are maintained, then both verbal and physical violence is reduced. If there is relapse, then the rate of violence increases again.
Evidence-Based Policies to Reduce Alcohol-Related Violence8
Alcohol creates a wide range of negative consequences that have implications in the public health policy realm. Toben Nelson observed that while several behaviors are risky, not all of them are public health issues because they do not result in population-level harm. Alcohol consumption, on the other hand, is a common social activity that is associated with several risks to the public as a whole. The risks associated with alcohol consumption increase as exposure increases. However, consumption of alcohol lies on a curve, with a larger number of the population at the lower consumption end, and a smaller number at the higher end. Thus, even though high consumption carries a greater risk to individuals, lower levels of consumption, which occur at greater frequency, pose a greater risk to the population as a whole. As epidemiologist Geoffrey Rose noted, “a large number of people exposed to a small risk may generate many more cases than a small number exposed to a high risk.” The vast majority of health harms in a community arises from a moderate or lower level of risk.
This yields two approaches to harm reduction, Nelson noted: the high-risk approach that provides individual-level treatment, and a population-level
8 This section summarizes information presented by Toben Nelson, University of Minnesota School of Public Health.
approach that aims to change behavior and the conditions that shape that behavior. Often, these two approaches are seen as oppositional, when they should be complementary. Nelson spoke of a theory developed by Alex Wagenaar and Cheryl Perry that looked at the relationship of high-risk individuals embedded within a community. While much research has focused on drinking and alcohol-related problems or individual risk factors that increase drinking behavior, Wagenaar and Perry considered that, within the community, there is a wide range of availability of alcohol. Problems of alcohol at the individual level are a function of the economic, legal, or physical availability of alcohol in those communities. This availability, in turn, is shaped by policies and norms around how alcohol is provided or restricted. Nelson and his colleagues examined the efficacy of some of these policies around the United States and created a list of the top 10:
- Alcohol excise taxes (state)
- State alcohol control systems (monopoly)
- Bans on alcohol sales
- Outlet density restrictions
- Wholesale price restrictions
- Retail price restrictions
- Alcohol beverage control agencies that are present, functional, and adequately staffed
- Dram shop liability laws
- Hours of sale restrictions
- Restrictions on alcohol consumption in public places and events
Regarding taxes, Nelson noted the evidence that price strategies are inversely related to violence: the higher the price of alcohol, the lower the rates of alcohol-related violence (Wagenaar et al., 2010). On alcohol retail density, studies show more violence and violent crime where there is greater density, and time-series data show increases in violence when alcohol outlets privatize and proliferate. Despite evidence for these types of programs being effective in reducing harm related to alcohol consumption, there has not been increased uptake of them, Nelson noted. Instead, policies judged to be less effective are on the rise in the United States.
Alcohol Policy: Challenges and Successes in Latin America9
In terms of the global burden of disease, alcohol has higher impacts in countries in Latin America than in the United States or Europe. In Brazil, an
9 This section summarizes information presented by Ronaldo Laranjeira, Universidade Federal de São Paulo and National Institute on Alcohol and Drug Policy, Brazil.
unregulated alcohol market and close ties between the alcohol industry and politicians are challenges for the alcohol control policy landscape. There is little awareness among policymakers and little community involvement in alcohol policy, and there are few good examples of successful alcohol policies and programs within Brazil.
Ronaldo Laranjeira further described the unregulated market in Brazil: There is no licensing requirement to sell alcohol, resulting in nearly 1 alcohol outlet per 200 people. In addition, an estimated 30 percent of drivers on weekends are intoxicated, and there is little restriction on adolescent purchase of alcohol. The price of alcohol is inexpensive as well. Laranjeira noted that one can of beer costs 30 cents (by comparison, a liter of orange juice costs $3.50). Alcohol is also marketed heavily in Brazil. Laranjeira stated that normally, sale of alcohol in sports stadiums is forbidden by law. However, for the World Cup in 2014, lobbying from both the International Federation of Associated Football (FIFA) and the alcohol industry succeeded in changing that law.
Despite these and other challenges, Laranjeira shared an example of a successful program in the city of Diadema, in São Paolo state. Diadema has around 350,000 inhabitants, most of whom are low to middle class. In the 1990s, there was a high rate of homicides—102 per 100,000 people—with 50 percent occurring between 9 p.m. and 6 a.m. There was also a high rate of violence against women during that time, as well as a high incidence of gang activity and car crashes. In 2002, a municipal law was passed prohibiting the sale of alcohol between 11 p.m. and 6 a.m. The law was enforced, with local police verifying compliance every night. The first violation resulted in a warning, the second in a fine, and the third in a fine and the working permit license suspended. Despite the previously mentioned lack of licensing, enforcement on this law was active, and there was high approval among the community.
Ten years of homicide data were examined, from 1995 to 2005, including homicide data both before and after the law was enacted. The results showed that 528 lives were saved, with a 46 percent reduction in homicides. Additional years of data have reinforced this reduction in violence, with a homicide rate of less than 20 per 100,000 people in the past few years and a decrease in violence against women, as well (Duailibi et al., 2007). Laranjeira closed by noting that one important factor in the success of this program was the continued enforcement; success of program replications in other cities has decreased because of a failure to sustain the nightly police checks.
In the ensuing discussion, speakers and workshop participants further explored some of the main themes raised, notably the importance
of enforcement in setting policy and the role of the unregulated alcohol markets and the alcohol industry. Participants also discussed additional policy approaches, such as those that restrict alcohol use and firearms possession, and interventions that merge population-based and individual-focused perspectives.
Andrés, R., and K. Hempstead. 2011. Gun control and suicide: The impact of state firearm regulations in the United States. Health Policy 101:95–103.
Bushman, B. J. 1997. Effects of alcohol on human aggression. Validity of proposed explanations. Recent Developments in Alcoholism 13:227–243.
Duailibi, S., W. Ponicki, J. Grube, I. Pinsky, R. Laranjeira, and M. Raw. 2007. The effect of restricting opening hours on alcohol-related violence. American Journal of Public Health 97(12):2276–2280.
Gunnell, D., M. Eddleston, M. Phillips, and F. Konradsen. 2007. The global distribution of fatal pesticide self-poisoning: Systematic review. BMC Public Health 7:357.
Ito, T. A., N. Miller, and V. E. Pollock. 1996. Alcohol and aggression: A meta-analysis on the moderating effects of inhibitory cues, triggering events, and self-focused attention. Psychological Bulletin 120(1):60–82.
Jacob, T., K. E. Leonard, and J. R. Haber. 2001. Family interactions of alcoholics as related to alcoholism. Alcoholism: Clinical and Experimental Research 25:834–843.
Leonard, K. E., and B. M. Quigley. 1999. Drinking and marital aggression in newlyweds: An event-based analysis of drinking and the occurrence of husband marital aggression. Journal of Studies on Alcohol and Drugs 60(4):537–545.
Leonard, K. E., and L. J. Roberts. 1998. The effects of alcohol on the marital interactions of aggressive and nonaggressive husbands and their wives. Journal of Abnormal Psychology 107(4):602–615.
Leonard, K. E., and M. Senchak. 1996. Prospective prediction of husband marital aggression within newlywed couples. Journal of Abnormal Psychology 105(3):369–380.
Miller, M., S. J. Lippmann, D. Azrael, and D. Hemenway. 2007. Household firearm ownership and rates of suicide across the 50 United States. Journal of Trauma and Acute Care Surgery 62(4):1029–1034.
Murphy, C. M., J. Winters, T. J. O’Farrell, W. Fals-Stewart, and M. Murphy. 2005. Alcohol consumption and intimate partner violence by alcoholic men: Comparing violent and nonviolent conflicts. Psychology of Addictive Behaviors 19(1):35–42.
Parrott, D. J., and P. R. Giancola. 2004. A further examination of the relation between trait anger and alcohol-related aggression: The role of anger control. Alcoholism: Clinical and Experimental Research 28(6):855–864.
Stevens, J. A., P. S. Corso, E. A. Finkelstein, and T. R. Miller. 2006. The costs of fatal and non-fatal falls among older adults. Injury Prevention 12(5):290–295.
Swanson, J. W., L. K. Frisman, A. G. Robertson, H. J. Lin, R. L. Trestman, D. A. Shelton, K. Parr, E. Rodis, A. Buchanan, and M. S. Swartz. 2013. Costs of criminal justice involvement among persons with serious mental illness in Connecticut. Psychiatric Services 64(7):630–637.
Van Dorn, R., J. Volavka, and N. Johnson. 2012. Mental disorder and violence: Is there a relationship beyond substance use? Social Psychiatry and Psychiatric Epidemiology 47(3):487–503.
Vigdor, E. R., and J. A. Mercy. 2006. Do laws restricting access to firearms by domestic violence offenders prevent intimate partner homicide? Evaluation Review 30(3):313–346.
Wagenaar, A. C., A. L. Toblet, and K. A. Komro. 2010. Effects of alcohol tax and price policies on morbidity and mortality: A systematic review. American Journal of Public Health 100(11):2270–2278.
Webster, D. W., J. S. Vernick, A. M. Zeoli, and J. A. Manganello. 2004. Association between youth-focused firearm laws and youth suicides. JAMA 292(5):594–601.
Wiebe, D. J. 2003. Homicide and suicide risks associated with firearms in the home: A national case-control study. Annals of Emergency Medicine 41(6):771–782.
This page intentionally left blank.