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

3. Overview of Firearm Deaths and Injuries (cont'd)

3. Overview of Firearm Deaths and Injuries (cont'd)

3.9 Costs of Firearm Injuries and Deaths

Research on the costs of injuries related to firearms has commonly taken one of two approaches: the first compares the relative cost-effectiveness of various types of medical intervention in the case of firearm injuries. This type of research is often called for but rarely conducted. For example, Ordog and colleagues, studying the costs and benefits of outpatient treatment versus hospitalization for gunshot wounds, reviewed the records of all patients discharged without hospitalization at the King/Drew Medical Centre in Los Angeles between 1977 and 1991. They found that 60 percent of patients had been treated as outpatients after an emergency department evaluation and treatment; all of the patients were considered to have minor gunshot wounds. The complication rate was fairly low at 1.8 percent, and most often involved an infection. The cost savings were estimated at $37 million U.S. (Ordog et al., 1994). Examples of such research were not found in Canada.

The second and most frequent type of research quantifies the medical or economic costs of firearm injuries for a country or a specific jurisdiction. Such costs include emergency transport and services; emergency and other medical care; burial; mental health care; loss of productivity; administration; and costs of pain, suffering and lost quality of life (Miller and Cohen, 1996: 49). Since detailed data on the specific costs of injuries is generally not available, researchers often have to make assumptions and rely on estimates to attribute a dollar figure to certain injuries. The estimates may vary broadly depending on the types of costs that researchers consider and the nature of the assumptions and calculations they have made. For a recent review of this literature, see Injury Prevention Centre Edmonton, 1996.

Some have questioned the usefulness of this type of research. Its purpose is often to express in financial terms the serious harm that can result when people misuse firearms. However, many studies of this type fail to compare the costs to those that result from other forms of injury (Mauser, 1996c: 5). It has been argued that, to be more useful, studies that assess the costs of firearm-related injuries would also have to consider that the majority of these are caused intentionally (Ibidem), leaving researchers little basis to assume that there would be no injuries and no resulting costs if firearms were not available.

The previous literature review (Gabor, 1994:15) quoted a few U.S. studies, but noted that these costs had not yet been systematically assessed in Canada. Researchers should be cautious about drawing conclusions from U.S. data and applying them to the Canadian situation (Gabor et al., 1996: 323). The two countries differ, sometimes greatly, in terms of health care systems and costs, the prevalence and nature of firearm injuries, the context in which they occur and the types of weapons that are commonly used. As well, many of the U.S. studies have methodological limitations and problems with data availability, not the least of which is that accurate data on the incidence and seriousness of non-fatal firearm injuries are generally missing (Max and Rice, 1993: 182; Kellermann et al., 1996: 1442). Max and Rice (1993: 183) concluded that estimating the costs of firearm injuries given available data really amounts to "shooting in the dark." The statement apparently remains valid, whether applied to the United States or to Canada.

Miller, taking a broad approach, attempted to estimate the total costs of gunshot injuries and deaths in Canada in 1991. Based largely on previous U.S. research, extrapolations from U.S. data, and various secondary data sources, Miller estimated that these costs totaled $6.6 billion, or $235 per capita (Miller, 1995), with the largest component reflecting lost quality of life. When the author listed the costs according to the intent of the shooter, suicides and attempted suicides topped the list at $4.7 billion, followed by homicides and assaults at $1.1 billion, and accidental shootings at nearly $602 million. The study was criticized for a number of reasons, including relying too heavily on extrapolations from U.S. data (Mauser, 1996c: 4-6; Nakamura, 1996; Rosenberg, 1996; Smart, 1996; Sobrian, 1996a; Suter, 1996; and a reply by Miller, 1996).

Recently, the Edmonton Injury Prevention Centre conducted a pilot project to collect primary data on the direct medical costs of firearm injuries in Alberta in a one-year period between 1993 and 1994. The Centre collected data from a written survey of hospitals and supplemented that information with secondary data. The study produced cost estimates based on the actual services for these injuries and the actual or estimated costs associated with these services. Excluding the costs of non-hospital physician visits, community rehabilitation visits for physiotherapy or other treatments, and medications or long-term care costs, the total direct medical costs of firearm injuries in Alberta in that period were $869,404. The cost of acute hospitalization services accounted for nearly 70 percent of the total estimated cost. In that province, hunting rifles were most often the cause of injuries requiring emergency room treatment and acute hospitalization. The study also revealed that the highest costs of treating individual firearm injuries were from self-inflicted firearm injuries and from shotgun injuries (Injury Prevention Centre Edmonton, 1996).

The pilot project also involved telephone interviews with officials from nine other provinces to find out if data on medical costs were available and to determine the feasibility of replicating the Alberta study. The authors concluded that the study could be replicated, but that researchers would encounter challenges similar to those found by the Centre, and that the results would likely suffer from the same limitations. They suggested that more accurate estimates could be produced, given sufficient resources, by identifying all firearm injuries in a given population and collecting primary cost data instead of relying on secondary data sources to estimate the relevant costs (Injury Prevention Centre Edmonton, 1996: 48).

3.10 Prevention Methods

In an article entitled The Role of the Health Community in the Prevention of Criminal Violence, Gabor, Welsh and Antonowicz (1996) proposed that people should consider fatal and non-fatal injuries resulting from violent incidents involving a firearm as a critical public health problem and as a threat to community health as opposed to community order. They argued that crime should be viewed within the wider context of health problems such as illnesses or accidental injuries and that risk factors associated with crime and victimization should be identified and addressed as early as possible, not just by the criminal justice system, but also by the health community (Gabor et al., 1996:324).

Many U.S. authors believe that injuries and deaths caused by firearms can be prevented, and that the most promising approach is through public health (Camosy 1996; Cohen and Swift, 1993; Elders, 1994; French, 1995; Goetting, 1995; Hargarten et al., 1996; Johnson, 1993; Kellermann, 1993; Kellermann et al., 1991, 1996; Lee and Harris, 1993; Mercy 1993; Powell et al., 1996; Roth, 1994, Teret and Wintemute, 1993; Weiss, 1996; Zwerling and Merchant, 1993; Zwerling et al., 1993). From that perspective, public health sources can provide primary and secondary prevention methods once the risk factors associated with firearm injuries are identified (Camosy, 1996: 971). That view follows the theory that the most cost-effective way to control disease is to prevent it from occurring (Kellermann et al., 1991: 19). Kellermann and his colleagues proposed ten strategies for preventing firearm injuries (Idem: 34-35). According to them, "experience with other public health interventions has shown that prevention is best accomplished by first identifying, then breaking, the chain of disease causation at its weakest link" (Idem: 19). They added that the weakest link may not always be obvious or proximate to the illness or injury (Ibidem).

Weiss (1996: 201) remarked that, within the public health or epidemiological model, violent behaviour is likely to follow a similar pattern to that of any public health epidemic. Kellermann and his colleagues noted that, although the strategies they proposed were developed as countermeasures to prevent unintentional injuries, they may apply to intentional ones as well (Kellermann et al., 1991: 21). Hargarten and colleagues expressed a similar view (Hargarten et al., 1996). Blackman, on the other hand, argues that this generalization to intentional injury is still very much an "untested assumption" (Blackman, 1996: 1273; see reply by: Hargarten et al., 1996a).

The public health approach has inspired many calls for comprehensive firearm injury prevention programs, and many of its proponents prefer strategies that will reduce the number of firearms that are available. Gabor stated that, "to achieve a significant effect on public safety, measures would have to achieve considerable reduction in the proportion of households with firearms" (Gabor 1996: 106). Similarly, Chapdelaine and Maurice argue that injuries, "always involve access to a firearm by a person who can discharge it" (1996: 1286). They added: "This access constitutes the universal link; the one against which we can take action in the chain of events leading to an injury from a firearm" (Ibidem). Others have said that even if the public health approach would dictate that the root causes of violence can be addressed, the lethality of firearms is such that reduced access to firearms and, in particular, to handguns is necessary to reduce the lethality of violence (Powell et al., 1996: 208).

3.11 Summary