Firearms, Accidental Deaths, Suicides and Violent Crime: An Updated Review of the Literature with Special Reference to the Canadian Situation

6. Accidents

6. Accidents

6.1 Frequency of Accidental Firearm Deaths

Over the last few decades, the rate of unintentional firearm deaths in Canada and most other industrialized countries has been declining steadily. In Canada, that decline actually began in the 1950s, and was particularly evident in the 1960s and early 1970s (Kopel, 1992; Mauser, 1995a). Steady decreases in the rate of unintentional deaths were also reported in many other countries, including Denmark (Thomsen and Albrecktsen, 1991: 166), Australia (Mukherjee and Carcach, 1996: 8-9) and the United States (Jacobs, 1995: 325; Kates et al., 1995; Kleck, 1991; Lee and Harris, 1993: 16).

In 1995, 49 people died in Canada because of an unintentional firearm injury. This represents about four percent of the 1,125 firearm-related deaths reported that year. In other countries, similar data are not always reliable, making international comparisons difficult. According to the United Nations Survey on Firearm Regulation, the rate of accidental deaths resulting from a firearm, per 100,000 population, was 0.02 in the United Kingdom, 0.11 in Australia, 0.13 in Canada, 0.29 in New Zealand, and 0.58 in the United States (United Nations, 1998: 108-109).

The case-fatality rate of unintentional injuries can probably be assumed to be smaller than the case-fatality rate of intentional firearm injuries. There is limited information on which to base a valid Canadian estimate of the case-fatality rate for accidental firearm injuries. When the number of patients admitted to the hospital for an accidental firearm injury and discharged after one night was compared to data on the number of people who died from their firearm injuries, the data suggest that unintentional injuries are ten times more frequent than unintentional deaths (Gabor 1995: 205). In the past several years, this gap has been increasing. In the most recent year for which data is available (Hung, 1997), there were 13 times more unintentional injuries than unintentional deaths.

6.2 Characteristics of the Individuals and Circumstances Involved

Relatively little is known about the characteristics and circumstances of firearm accidents in Canada. Existing national data provide information on the number of people who die each year from unintentional firearm injuries; the age and sex of the victim; and the jurisdiction in which the person died. A few coroner’s offices have conducted some preliminary research on these deaths. The provincial departments of Natural Resources collect some data on firearm accidents, but they are generally limited to hunting incidents and do not provide sufficient detail about the individuals or circumstances involved.

Canadian research is limited on the circumstances of firearm accidents, such as hunting and other types of accidents, and individual and environmental factors. Researchers recently conducted a study in Quebec and found that 37 percent of accidental deaths resulted from a hunting incident; five percent occurred when the shooter was carrying a firearm; and 48 percent occurred during other activities. In 55 percent of cases, the shooter accidentally shot himself. In half of the incidents, the shooter owned the firearm (Bureau du coroner, 1994: 64-66), implying that half the time the incident involved a borrowed or stolen weapon. The Quebec study also found that 95 percent of the accident victims were male (Bureau du coroner, 1994: 64-66).

Gabor (1995: 205) reported that, between 1979 and 1988, 25 percent of people who died from an unintentional firearm injury were under the age of 15, and 30 percent of victims were between the ages of 15 and 24. The Quebec coroner’s office recently conducted a study of 38 unintentional firearm deaths that occurred in the province between 1990 and 1992. The findings indicated that, of these, 13 percent of the victims were younger than 15, and 39 percent were between 15 and 24 years old. When the researchers examined the per capita rates, they found a bi-modal distribution: the highest rates were found in the 15 to 24 age group and the second highest in the 55 and older group (Bureau du coroner, 1994: 11).

The U.S. Centers for Disease Control and Prevention (1997: 103) examined the causes of death in 26 developed countries among children up to 14 years of age, using national health statistics provided by the countries. The research showed that the death rate for children who were victims of unintentional firearm injuries was nine times higher in the United States–0.36 compared to 0.04–than in all of the other countries combined.

According to one national study in the United States, seven male children died from firearm injuries for every female child victim in 1989. The ratio increased to 15:1 among adolescent and young adult victims (Lee and Harris, 1993: 17). Children aged 10 to 19 and living in non-metropolitan areas were twice as likely to die from an unintentional firearm injury (Ibidem). This trend may depend on whether victims had easy access to emergency or hospital care in the areas of the study.

The Cook County Medical Examiner’s Office in the United States holds records on 45 children under the age of 10 who died from firearm injuries between 1984 and 1992. Choi and colleagues reviewed all of these cases (1994). Fourteen, or 31 percent of these were accidents resulting from children playing in the home; the remaining 69 percent were homicides. Approximately 53 percent of the children were younger than six when they were killed, and 78 percent of the victims were male. Handguns were used in 89 percent of cases. No comparable studies were found for Canada.

6.3 Predisposition to Accidental Injuries

In the previous literature review, Gabor (1994:56) considered whether people who are injured or killed because of firearm injuries are part of a highly predisposed group or whether the injuries and deaths are owed to the dangerousness of firearms. He concluded that considerably more research is required to establish whether those involved in firearm accidents are reckless or merely unfortunate individuals. He added that "the existing evidence suggested that although reckless individuals may be more likely than others to be involved in accidents, the majority of cases involve product design shortcomings or ordinary people–often young people–who had access to a firearm and made an error in judgement or were the victims of misfortune" (Idem: 58). Since then, there has been no research to shed new light on this question.

6.4 Children and Youths Injured in Firearm Accidents

The previous review raised a point concerning the age of the victims of firearm accidents (Gabor, 1994: 53). The author noted how some supporters argued that the relative youthfulness of victims who die from an accidental firearm injury should incite greater restrictions on firearms, while others insisted that the number of fatal firearm accidents involving children is small in comparison to other causes of death for children and that it should not be used to justify further firearm restrictions (Gabor, 1994: 54).

When researchers examined Connecticut hospital records of children up to age 19 who were treated for gunshot wounds between 1988 and 1992, they found that more victims had suffered non-fatal injuries that were classified as accidental or were undetermined than had died under these circumstances. In total, these two categories accounted for only seven percent (6 percent and one percent respectively) of fatal incidents, compared to 57 percent (39 percent and 18 percent respectively) of the non-fatal cases (Zavoski et al., 1995: 279). In contrast, twice as many incidents in which a child had died were attributed to assault than ones in which the child was injured but did not die (81 percent and 41 percent respectively). About 25 non-fatal incidents were motivated by suicide for every one in which the child died (Ibidem). These data may lead to questions about how firearm incidents involving children, particularly the non-fatal incidents, were being classified and how many of them were truly accidental.

Li et al. (1996) analyzed data from the National Pediatric Trauma Registry for the period between 1990 and 1994. They examined patients 14 years of age and younger who were admitted to trauma centres for unintentional firearm injuries (n=292) and assaultive firearm injuries (n=457). The researchers found the frequency of unintentional firearm injuries rose in the afternoons peaking between 4:00 and 5:00 p.m., with 89 percent having occurred at home. Over 80 percent of the patients were male, and 58.9 percent were between 10 and 14 years of age. The study concluded that unintentional firearm injuries involving children occurred while parents were still at work and children were playing with loaded guns.

6.5 Firearm Injury and Prevalence of Firearms

Very few studies have probed how accidents caused by firearms may be linked to the availability of firearms (Gabor, 1994: 55;1995: 205; Mayhew, 1996:23). The author of the previous review concluded that, in his opinion, "the preponderance of evidence suggests that increasing the number of firearms in circulation will lead to more fatal accidents" but that more research was required before researchers could draw definitive conclusions (Gabor, 1994: 56). Some Canadian researchers argue that studies have often demonstrated a strong link and that national comparisons bear out the relationship between the number of people who own firearms and the number who die from unintentional firearm injuries (Gabor et al., 1996: 324). Others suggest that a number of factors affect the relationship(e.g., Kopel, 1995). Data collected from about 20 countries that participated in a recent United Nations survey and that had information on the levels of firearm ownership and on the rates of accidental deaths caused by firearms, suggest that there may be a relationship between the two (United Nations, 1997b: 36). However, as with previous research, one cannot determine if this is a causal relationship because of other variables that come into play.

The lack of measures on how accessible firearms are in countries has restricted comparisons of international data (Mayhew, 1996:23; Lester, 1993a: 167). None of the studies referenced above proposed an international comparison of data on people who were injured by firearms and data on people who were killed by them. Until more is known about the prevalence of non-fatal injuries suffered from firearm accidents, researchers are unlikely to answer the question satisfactorily.

6.6 Prevention Strategies

6.6.1 Safe Storage

Loaded firearms within reach are quite possibly responsible for a number of spontaneous homicides, suicides and accidents, particularly involving children (Morrison et al., 1995: 364). Many authors have argued that preventing easy access to a loaded firearm can reduce the risk of firearm accidents. In the United States, many see preventing a child’s access to a handgun as the first step in preventing firearm morbidity and mortality against children (Goldberg et al., 1995; Laraque et al, 1995; Senturia et al., 1996; Wiley and Casey, 1993; Zavoski et al., 1995: 281). In support, Kleck argued that in none of the studies he had reviewed where the firearm was kept under lock, was a child killed in an accident involving a firearm (Kleck, 1991: 279). There have been various strategies proposed to reduce firearm accidents, including firearm safety courses, laws mandating safe storage of guns, restrictions on who may buy guns, educating children about the dangers of firearms, and strategies that focus on the design of firearms. These measures and others may be able to reduce firearm injuries and deaths among children and adults.

Several U.S. studies have examined how owners store their firearms. National random telephone surveys of firearm owners (Hemenway et al., 1995a; Cook and Ludwig, 1997) found that as many as 20 percent of owners kept a loaded firearm unlocked in the home (Hemenway et al., 1995a: 49; Cook and Ludwig, 1997: 7). A 1991 survey indicated that as many as 53 percent of the respondents did not keep firearms locked up. Owning a handgun as opposed to a long gun, purchasing the weapon for protection, and having no children in the household were all positively correlated with keeping a loaded gun in the home. The strongest predictor of respondents keeping loaded firearms, however, was the type of gun. Handgun owners were up to five times more likely than owners of long guns to keep the firearms loaded at least some of the time (Cook and Ludwig, 1997: 7; see also: Hemenway et al., 1995). Two telephone surveys of firearm owners in Oregon, conducted in 1991 and 1992, strongly suggested that unsafe firearm carrying and storage practices were associated with alcohol consumption patterns (Nelson et al., 1996).

Relevant U.S. studies confirm that when a person owns a handgun for work or to protect himself, this strongly indicates that a firearm is likely to be found loaded in the home, even in a home with children (e.g., Goldberg et al., 1995: 160; Senturia et al., 1996: 268; Morrison et al., 1995: 364). A survey of 102 police departments in cities with a population of 10,000 or more was conducted. In this study, an interviewer, posing as a parent of a three-year-old and a ten-year-old telephoned for advice on how to store a firearm safely (Denno et al., 1996). The interviewer asked the respondents about their own storage practices at home and discovered that trigger locks were frequently recommended but infrequently used by police officers themselves; portable lock-boxes were also recommended and were more frequently used by officers (Idem: 929). The researchers noted that "there was a remarkable contrast between what police recommended to the interviewer and what they used themselves. Over one-third of respondents reported using no storage method at all" (Ibidem).

6.6.2 Safety Training

Training firearm owners and prospective owners in the proper use and storage of firearms is often cited as a preferred method of preventing firearm accidents (e.g., Becker et al., 1993: 282). The previous literature review suggested that the provisions for the safe storage of firearms, introduced by the Canadian legislation in 1991, could play a role in preventing some tragedies (Gabor, 1994: 58). In 1991, a national survey showed that in 49 percent of the households in which at least one firearm was owned, at least one person took training on firearm safety during the previous five years (Angus Reid, 1991). More recent data on the number of individuals who took such training were not available nor was there any new Canadian research that examines the effectiveness of such strategies in reducing firearm accidents.

Some safety programs are designed to teach children about the risks of firearms. An experimental U.S. study of 48 children, aged four to six years, compared children’s play with toy guns and disarmed real handguns before and after an information-based intervention. Its results indicated that the intervention did not change the children’s behaviour (Hardy et al., 1996). Many of the children had difficulty differentiating real guns from toy guns. While children who had access to a firearm at home were better able to tell the difference, some of these children too were not aware that they had just played with a real handgun (Ibidem). The researchers concluded that only providing children with information is insufficient; parents must be responsible for protecting their children from the potential hazards of firearms.

6.6.3 Safety of Firearms

Improvements to the safety of firearms are often recommended as a way to prevent accidental injuries and deaths. Chapdeleine and colleagues noted that self-inflicted accidental injuries can be related to faulty firearm design and poor maintenance. Many of them are preventable. A better design–one focused on preventing children and adults from inadvertently discharging a firearm–may have a greater impact than years of safety training (Chapdeleine et al., 1991:1220).

In the United States, several authors have advocated a product-safety oriented focus as part of a larger firearm injury prevention approach (e.g., Donzinger, 1996; Kellermann et al., 1991; Marwick, 1995; Sinauer et al., 1996; Wintemute, 1996). Often noting the significant progress made in reducing the rate of motor vehicle deaths by improving the design of motor vehicles, they emphasize the need to identify those trends in the design and marketing of firearms that affect the likelihood of firearm injuries or appear likely to do so in the future (Wintemute, 1996: 1749). Donzinger argued that technology has reached a point where "fingerprinted" weapons are made that can be fired only by the legal owner (1996: 214). On the question of firearm design and safety features, some authors noted the importance of providing protection that does not depend on users’ behaviour (Sinauer et al., 1996: 1743). They argued that design modifications such as loading indicators, or minimum standards for trigger-safety mechanisms, could reduce the likelihood of unintended discharges (Ibidem). While some of these technological advances have been made, they are not widespread and there does not appear to be any research on their effectiveness.

6.6.4 Hunting Regulations

Other specific measures are aimed at preventing hunting accidents. These are often implemented as part of hunting and gaming laws and regulations. Researchers have evaluated the Hunter Orange law in North Carolina, and found that fewer hunters were being killed because they were mistaken for game. These numbers were statistically significant (Cina et al., 1996: 395). No Canadian studies have examined how wearing blaze orange hunting gear can reduce hunting accidents.

6.7 Summary