Effect of current federal regulations on handgun safety features☆☆☆★

Introduction 

Firearm injuries are a major cause of premature death and disability in the United States. In 1999, approximately 824 US citizens died of unintended gunshot wounds, with the vast majority caused by handguns.1, 2 On average, 13,688 individuals sustain unintentional nonfatal injuries each year, with almost 60% of those caused by handguns.3, 4

Handguns are a common consumer product in homes throughout the United States. If not handled and stored carefully, they can cause unintended death or injury.4, 5 Historically, efforts to prevent unintended handgun injuries have focused on firearm safety education and the promotion of safe storage.6 Little attention has been given to designing safer handguns.7

Over the past half century, the rate of automobile fatalities has been substantially reduced by changing the way cars and roadways are designed, as well as by aggressive efforts to modify the behavior of drivers through education and law enforcement.7, 8 Safety improvements, such as seat belts, air bags, a third brake light, and safety cage construction, have helped reduce the morbidity and mortality associated with automobile crashes. Design changes of this sort are consistent with product safety engineering theory, which determines the type of injuries that occur with a consumer product and makes modifications to prevent or mitigate those injuries.9

Over the past 125 years, as handgun designs have become more refined, engineers have sporadically included safety devices in an attempt to address specific safety concerns. For example, a grip safety was added to Smith & Wesson revolvers in 1889 to make the gun difficult for children to fire.10 Magazine safeties were included on Remington pistols in 1915 to “prevent firing the forgotten cartridge.”10 Walther included loaded chamber indicators on several of its models so that users would always know whether the chamber contained a cartridge. 11 Transfer bars, patented by Iver Johnson Arms in 1896, were created to prevent discharge if the gun is unintentionally dropped on its hammer.12 Most recently, Smith & Wesson, the largest producer of handguns in the United States, has begun including built-in locks on many of its revolver models.13 Although the effectiveness of these devices has not been formally evaluated, the lessons learned from other consumer products suggest that it should be possible to build safer handguns.14

In 1968, Congress passed the Gun Control Act, which barred the importation of handguns that are not “generally recognized as particularly suitable for or readily adaptable to sporting purposes.” 15 The primary intent of this act was to reduce the importation of inexpensive, easily concealed handguns, which were believed to be disproportionately involved in crime.16, 17, 18 The goal of improving the safety and quality of handguns, although noted at the time, was of secondary importance.16, 17, 18 The Bureau of Alcohol, Tobacco, and Firearms (ATF) developed a set of design and performance specifications known as the “factoring criteria” to implement this “sporting purposes” test.19 Under the factoring criteria, manufacturers have an incentive to incorporate several intrinsic passive safety features into pistols. Also, all revolvers are required to pass a performance-based drop-safety test.

Domestic manufacturers were exempted from the factoring criteria, but in recent years, federal legislation has been proposed that would extend the factoring criteria to all handguns sold in the United States, whether manufactured domestically or internationally.20, 21, 22, 23 Additionally, some state and local governments, including Maryland, Massachusetts, and California, have passed legislation mandating safety standards for all handguns sold in those jurisdictions. 24, 25, 26 Supporters of this legislation hope, in part, that extending the factoring criteria will improve handgun safety by giving gun manufacturers a stronger incentive to include safety features in their products.27

To assess the potential effect of extending the factoring criteria to domestic handgun designs, we compared handguns manufactured in the United States with those imported from other countries. We were particularly interested in determining whether imported handguns, which must comply with the factoring criteria, are more likely to include a loaded chamber indicator, a grip safety, a magazine safety, or a drop safety than domestically manufactured handguns. We also were interested in determining whether domestic handguns that already meet the factoring criteria are more likely to contain these passive safety devices than handguns that fail to meet the criteria.

Methods 

The factoring criteria are a standardized set of requirements, published on ATF Form 4590 (5330.5),19 that imported handguns must meet to be sold in the United States (Tables 1 and 2). To meet the factoring criteria, a handgun must first satisfy several prerequisites and then achieve a passing score on tests of individual characteristics. The prerequisites for pistols require the gun to have a manually operated safety and have a combined height and length of at least 10 in (25.4 cm), with the height being at least 4 in (10.2 cm) and the length being at least 6 in (15.2 cm). A pistol that meets these prerequisites is then scored on its total length, frame construction, weight, caliber, and the presence of specific safety features and other equipment. The pistol must achieve a minimum score of 75 points to pass. There is no maximum obtainable score.

Table 1. Text of the ATF factoring criteria, ATF Form 4590 (5330.5): Pistol.19 Adapted from Department of the Treasury, Bureau of Alcohol, Tobacco, and Firearms. Factoring Criteria for Weapons. Form 4590 (5330.5). Washington, DC: Bureau of Alcohol, Tobacco, and Firearms; 1983.

	Criteria
	Prerequisites
	1. The pistol must have a positive manually operated safety device. 2. The combined length and height must not be <10 in, with the height (right angle measurement to barrel without magazine or extension) being at least 4 in and the length being at least 6 in.
	Individual characteristics	Point Value
	Overall length (for each ¼ in longer than 6 in)	1
	Frame construction
	Investment cast or forged steel	15
	Investment cast or forged HTS alloy	20
	Weapon weight with magazine (unloaded), per ounce	1
	Caliber
	.22 Short and .25 auto	0
	.22 LR and 7.65 mm to .380 auto	3
	9-mm Parabellum and over	10
	Safety features
	Locked breach mechanism	5
	Loaded chamber indicator	5
	Grip safety	3
	Magazine safety	5
	Firing pin block	10
	Miscellaneous equipment
	External hammer	2
	Double action	10
	Drift-adjustable target sight	5
	Click-adjustable target sight	10
	Target grips	5
	Target trigger	2
	Passing score	75 points

HTS, High-tensile strength; LR, long rifle.

Table 2. Text of the ATF factoring criteria, ATF Form 4590 (5330.5): Revolver.19 Adapted from Department of the Treasury, Bureau of Alcohol, Tobacco, and Firearms. Factoring Criteria for Weapons. Form 4590 (5330.5). Washington, DC: Bureau of Alcohol, Tobacco, and Firearms; 1983.

	Criteria
	Prerequisites
	1. Must pass safety test. 2. Must have overall minimum frame (with conventional grips) length (not diagonal) of 4.5 in. 3. Must have a barrel length of at least 3 in.
	Individual characteristics	Point Value
	Barrel length
	<4 in	0
	For each ¼ in longer than 4 in	1
	Frame construction
	Investment cast or forged steel	15
	Investment cast or forged HTS alloy	20
	Weapon weight (unloaded), per ounce	1
	Caliber
	.22 Short to .25 ACP	0
	.22 LR and .3 to .38 S&W	3
	.38 Special	4
	.357 Mag and over	5
	Miscellaneous equipment
	Adjustable target sight (drift or click)	5
	Target grips	5
	Target hammer and trigger	5
	Passing score	45 points
	Safety test
	A double-action revolver must have a safety feature that automatically (or, in a single-action revolver, manually) causes the hammer to retract to a point at which the firing pin does not rest on the primer of the cartridge. The safety device must withstand the impact of a weight equal to the weight of the revolver dropping from a distance of 36 inches in a line parallel to the barrel on the rear of the hammer spur a total of 5 times.

ACP, Automatic Colt pistol; S &W, Smith & Wesson; Mag, Magnum.

The prerequisites for revolvers require the gun to have a device that prevents discharge if the gun is dropped on its hammer from a height of 36 in (91.4 cm) and to have a frame length of at least 4.5 in (11.4 cm) with a barrel length of at least 3 in (7.6 cm). A revolver that meets these prerequisites is then scored on its barrel length, frame construction, weight, caliber, and presence of other equipment. A revolver must achieve a minimum score of 45 points to pass. There is no maximum obtainable score.

The unit of analysis for this study was a handgun model. There is, however, no standard industry definition of a model. Some manufacturers market only a handful of models with a large number of variations in caliber, barrel length, size, grip, finish, or other features within each model. Other manufacturers choose to market each variation as a unique model. Variation in some of these features, particularly barrel length, sights, grip, and safety devices, will affect how a model scores on the factoring criteria (Table 1, Table 2). To reflect the effect of these variations on a model's performance against the factoring criteria, we considered model variants, for the purpose of this analysis, as unique models. Specialty competition and commemorative handguns were excluded from our sample because they are not believed to be representative of the bulk of handguns manufactured and sold in the United States.

The major domestic producers of both revolvers and pistols were determined on the basis of 1996 handgun production data obtained by written request from ATF. All models from pistol manufacturers with annual total production of greater than 10,000 guns (n=16 manufacturers) and revolver manufacturers with annual total production of greater than 25,000 guns (n=5 manufacturers) were included. This represented approximately 92% of US pistol production and 98% of US revolver production in 1996. The models produced in 1996 by these major US manufacturers were identified from Gun Digest ,28 a widely used listing of firearms in the consumer market. For imported pistols, no manufacturer-specific importation data were available, and therefore, all imported models from the 21 foreign manufacturers listed in Gun Digest 28 were included. No imported revolver models were included for reasons described later. This sampling technique generated 755 individual handgun models (477 domestic models and 278 imported models) from a total of 42 manufacturers.

Domestic pistol models were evaluated for the presence or absence of 4 passively acting safety devices (Table 3 and Figure 1), as well as their overall performance against the ATF factoring criteria.

Table 3. Devices for the potential reduction of unintended handgun injuries.

	Device	Description
	Loaded chamber indicator	A device, often a small pin or other indicator, that protrudes from the gun when a cartridge (the brass case containing the bullet, gunpowder, and primer) is in the chamber. If the user is able to recognize the indicator, this device might prevent unintentional discharges in situations in which the user was not otherwise aware the gun was loaded. The problem with loaded chamber indicators is that they are often difficult to recognize and have the potential to break off, potentially causing the user to incorrectly assume that a cartridge is not in the chamber.11, 29, 30, 36, 37
	Grip safety	A lever located on the grip of the gun that, when not compressed by the hand of the user, automatically secures the handgun. Originally these were designed as a means of child proofing handguns because small children would have difficulty compressing the lever while simultaneously pulling the trigger. A grip safety might also reduce the risk of unintended discharge when the gun in not being held by the grip, such as when handling it or showing it to another individual.10, 29, 30, 36
	Magazine safety	A device that prevents the pistol from discharging when the magazine is removed. If a pistol is always handled with the magazine removed until the user is ready to shoot, then a magazine safety might reduce the risk of unintended discharge. Also, if the gun and the magazine are stored separately, it might prevent unintended discharge by a child who finds the gun. However, an inexperienced user or a child might insert an empty magazine into the pistol, disengaging the safety, and mistakenly assume that the gun is unloaded. Also, there might be some combat-type situations in which a magazine safety could hinder the performance of the pistol if the magazine were dislodged.10, 30, 36, 37
	Drop safety (firing pin block)	A device that prevents the firing pin from unintentionally contacting the cartridge primer in cases in which the gun is dropped. Firing pin blocks are often found on pistols, on which they lock the firing pin in place until the trigger is pulled. Many revolvers have a transfer bar that inserts between the hammer and the firing pin only when the trigger is pulled, filling a gap that normally exists between the two.12, 36, 38

View Large Image Download to PowerPoint

Fig. 1. Diagram of pistol safety devices: A , loaded chamber indicator; B , grip safety; C , magazine safety; D , firing pin block.

The specific features of each model were determined by one of the authors (JSM) after receiving instruction on the application of the factoring criteria by the ATF. A sample of each model was obtained from the firearms collection of the Wisconsin State Crime Laboratory and inspected for the presence or absence of safety devices, as well as the other elements listed in the factoring criteria. These findings were compared with information provided in Gun Digest 28 and the Encyclopedia of Pistols and Revolvers .29 Data on frame composition and the passage of the drop test were obtained directly from the ATF after written request. Insufficient data were available on 6 domestic revolver models and 21 imported pistol models, and these were excluded from the study. However, all pistol models listed in Gun Digest 28 from the selected domestic manufacturers were included. Once this process was complete, the data were sent to the ATF for independent verification of accuracy.

Multivariate analysis was performed by using logistic regression. Three separate models were used to compare imported pistols with all domestic pistols, imported pistols with passing domestic pistols, and passing domestic pistols with failing domestic pistols. The dependent variable in the logistic equation was the group to which the gun belonged (ie, imported versus domestic), with the dummy variables for the presence of the 4 safety devices entered as independent variables. Differences were assessed by using odds ratios (ORs) and 95% confidence intervals (CIs). Goodness of fit was assessed by using the Hosmer-Lemeshow test. All statistical analysis was done with SAS statistical software (version 8.0, SAS Institute, Inc., Cary, NC).

Results 

A total of 755 handgun models were studied, of which 263 were domestic pistol models, 278 were imported pistol models, and 214 were domestic revolver models.

As required by federal law, all 278 imported handgun models met the factoring criteria. Of the 477 domestic handgun models studied, 316 (66%) met the ATF factoring criteria. This included 152 (71%) of the domestic revolver models and 164 (62%) of the domestic pistol models.

All but 2 of the domestic revolver models we studied would have passed the drop safety test included in the ATF factoring criteria (n=212 [99%]). All of these models contain one of a variety of safety devices to prevent unintentional discharge if the revolver is dropped. No revolvers in our sample included a grip safety, although historically, some revolver designs have included this type of device. Magazine safeties and loaded chamber indicators are not applicable to revolvers. Of the 66 revolver models that did not meet the factoring criteria, 52 (84%) did not meet the preliminary requirements, of which 46 (88%) failed because of a barrel length shorter than 3 in.

Of the 263 domestic pistol models studied, 33 (13%) contained a loaded chamber indicator, 53 (20%) contained a grip safety, 54 (21%) contained a magazine safety, and 183 (70%) contained a firing pin block (Table 4).

Table 4. Prevalence of safety devices and adjusted ORs for the inclusion of safety devices in imported versus domestic semiautomatic pistol models.

	Item	Imports (Total Models=278), No. (%)	Domestics (Total Models=263), No. (%)	OR (95% CI)*
	Loaded chamber indicator	62 (22)	33 (13)	1.59 (0.98–2.56)
	Grip safety	16 (6)	53 (20)	0.31 (0.17–0.58)
	Magazine safety	60 (22)	54 (21)	0.93 (0.62–1.47)
	Firing pin block	243 (87)	183 (70)	2.43 (1.54–3.85)
	*Hosmer-Lemeshow goodness-of-fit test for significant variables, P =.240.

Compared with domestic pistol models, imported pistol models were significantly more likely to contain a firing pin block (OR 2.43; 95% CI 1.54 to 3.85) and were also slightly more likely to contain a loaded chamber indicator (OR 1.59; 95% CI 0.98 to 2.56; Table 4). However, imported pistols were less likely to contain a grip safety (OR 0.31; 95% CI 0.17 to 0.58). The odds of containing a magazine safety were not statistically different between imported and domestic pistol models (OR 0.93; 95% CI 0.62 to 1.47).

Domestic pistol models that would meet the factoring criteria if applied (62% of all domestic pistol models) were just as likely to contain a loaded chamber indicator (OR 0.90; 95% CI 0.54 to 1.49), a magazine safety (OR 0.84; 95% CI 0.50 to 1.40), or a firing pin block (OR 1.40; 95% CI 0.80 to 2.46) as imported pistol models (Table 5).

Table 5. Prevalence of safety devices and adjusted ORs for the inclusion of safety devices in imported versus passing domestic pistol models.

	Item	Imports (Total Models=278), No. (%)	Passing Domestics (Total Models=164), No. (%)	OR (95% CI)*
	Loaded chamber indicator	62 (22)	31 (19)	0.90 (0.54–1.49)
	Grip safety	16 (6)	48 (29)	0.15 (0.08–0.29)
	Magazine safety	60 (22)	30 (18)	0.84 (0.50–1.40)
	Firing pin block	243 (87)	125 (76)	1.40 (0.80–2.46)
	*Hosmer-Lemeshow goodness-of-fit test not applicable because of only one significant variable.

Compared with domestic pistol models that did not meet the factoring criteria, domestic pistol models that passed were more likely to contain a loaded chamber indicator (OR 12.05; 95% CI 2.74 to 53.02), a grip safety (OR 24.12; 95% CI 7.83 to 74.33), or a firing pin block (OR 4.92; 95% CI 2.35 to 10.29; Table 6). Magazine safeties were more prevalent in pistols that met the criteria (OR 1.80; 95% CI 0.90 to 3.60), but this difference was not significant.

Table 6. Prevalence of safety devices and adjusted ORs for the inclusion of safety devices in passing domestic versus failing domestic pistol models.

	Item	Passing Domestics Total Models=164), No. (%)	Failing Domestics (Total Models=99), No. (%)	OR (95% CI)*
	Loaded chamber indicator	31 (19)	2 (2)	12.05 (2.74–53.02)
	Grip safety	48 (29)	5 (5)	24.12 (7.83–74.33)
	Magazine safety	30 (18)	24 (24)	1.80 (0.90–3.60)
	Firing pin block	125 (76)	58 (59)	4.92 (2.35–10.29)
	*Hosmer-Lemeshow goodness-of-fit test for significant variables, P =.995.

None of the pistols, neither imported nor domestic, contained all 4 safety devices studied. Only 9 (3%) of the imported models contained 3 safety devices, and none of the domestic models contained more than 2 safety devices. Two safety devices were found on 80 (49%) of the passing domestic models and 111 (40%) of the imported models. In comparison, only 10 (10%) of the failing domestic models contained 2 devices.

Discussion 

Automobiles and virtually every consumer product other than firearms are required to meet minimum federal and state standards for safety. No such federal standards exist for domestic handguns.

In general, few of the handgun models we studied contained passive safety devices. None of the pistol models we studied incorporated all 4 devices. Firing pin blocks, which reduce the risk of a gun discharging if it is dropped, were the most common device included in gun designs. Nearly 90% of imported pistol models and 70% of domestic pistol models had this feature. Virtually all of the domestic revolver models have a drop safety–type device, such as a firing pin block, and by definition, all imported revolvers have this type of device. However, none of the other safety devices were found on more than 22% of pistol models. Only one set of imported models, the Beretta Model 80 Cheetah Series, contained 3 safety devices (loaded chamber indicator, magazine safety, and firing pin block).

The factoring criteria were primarily implemented to restrict the importation of inexpensive, easily concealed handguns. However, the fact that the factoring criteria awards points for safety devices appears to have encouraged foreign handgun manufacturers to incorporate safety devices more regularly than their domestic counterparts. Yet, this effect is relatively modest. Most of the imported pistols do not contain a loaded chamber indicator, a grip safety, or a magazine safety. This is because the flexibility of the factoring criteria allows pistol models to pass without including any passive safety devices and only minimally rewards those that do. A maximum of 27 points of the 75 points required for a pistol to meet the factoring criteria can be obtained with safety devices. This number is modest when compared with the relatively large number of potential points obtainable on the basis of the size of the gun (1 point for every 1/4 in more than 6 inches in length and 1 point for every ounce of weight). On average, both imported and passing domestic pistols obtained approximately 45% of the total points earned for their size (weight and length) compared with 15% of their total points for safety devices.

We were surprised to note that imported pistol models were significantly less likely to contain a grip safety than domestic pistol models. This can be attributed to the influence of the Colt 1911A1 .45 caliber pistol, the standard-issue sidearm of the US Army for much of the 20th century.30 Variations of the popular 1911A1 design are currently produced by many manufacturers worldwide, with most incorporating the 1911A1's beaver tail–type grip safety. Of the 69 pistol models studied that contain a grip safety, both imported and domestic, 44 (64%) are .45 caliber pistols, with all of them being variants of the 1911A1. The majority of pistols in other calibers that contain a grip safety are also variations of the 1911A1 (n=22 [88%]). Fully 96% (n=66) of all pistol models that contain a grip safety, both imported and domestic, trace their roots to Colt's original design, with 77% (n=51) of these being domestic models.

It has been suggested that applying the factoring criteria to domestically manufactured handguns would improve the safety of these guns.31 Our results indicate that the universal application of the factoring criteria might lead to an increased prevalence of handgun safety devices, but the effect would be far from complete. The factoring criteria have been applied to imported handguns for more than 30 years, and yet a majority of imported pistols do not contain more than one passive safety device. This is because the factoring criteria are weighted to give a greater reward to large, powerful pistols than to those models that incorporate passive safety features designed to prevent unintended injuries. In contrast, the performance-based drop test for imported revolvers has led to the universal inclusion of drop safety devices on imported revolvers and might have contributed to the widespread prevalence of these devices in domestic revolver designs. We believe that these types of mandatory performance-based safety standards for all handguns would be the most effective strategy to increase the prevalence of safer guns, while giving manufacturers maximum design flexibility. The efficacy, however, of such a strategy for reducing firearm injuries is unknown.

Injury-control experts claim that unintended firearm injuries could be significantly reduced by the widespread use of well-designed safety devices, such as personalized handguns.31 Others argue that safety devices increase the complexity of gun designs and increase the risk of mechanical failure. Some fear that including safety devices in handgun designs might make owners more complacent and less attentive to safe handling practices. The effect of safer handgun design and safety devices needs further evaluation.

It might be possible, however, to extrapolate the potential effect of handgun safety engineering by considering what has been accomplished with automobiles.7 Since the mid-1960s, fatalities to occupants of motor vehicles have decreased, at least in part, because of the increased safety of the product.7, 32 Safety devices that exert their protective effect automatically, without active use, have made cars much safer than they once were. Performance-based standards for automobile safety devices are administered by the National Highway Traffic Safety Administration and have led to the development of safer new car designs.33

A 1991 General Accounting Office retrospective study suggests that intrinsic handgun safety devices could be effective at reducing the risk of unintended handgun injuries.34 However, no prospective data exist that examine the efficacy of the different types of safety devices in preventing injury or of how the different devices compare with each other. More data and analysis regarding the specific circumstances of unintended handgun injuries and the characteristics of the guns involved are needed to answer this important question.

The exact economic effect of including passive safety devices on handguns is unknown. Grip safeties, for example, can be found on some of the least expensive handguns, as well as on some of the most costly. Clearly the mere inclusion of a grip safety does not directly affect the retail price of the gun. The actual cost of including passive safety devices in handguns, however, when viewed in concert with any additional design, material, manufacturing, and marketing costs, will likely vary across the industry.

Our study is subject to important limitations. First, our analysis is based on the frequency of design features by model and not total units sold. We were unable to adjust our analysis to reflect the number of handguns of each model that were actually produced because the ATF is prohibited from collecting this information.

The results of this study are influenced by our definition of a handgun model. Because no standardized definition presently exists (each manufacturer selects how it will define its own models), we adopted a narrow definition to standardize application of the factoring criteria to handguns from all manufacturers. Handguns with only slight variations in design, such as differences in barrel length or caliber, were considered as unique models, irrespective of whether the manufacturer described them that way. Our estimate of the prevalence of magazine safeties and loaded chamber indicators is comparable with that seen in studies with a broader definition.35

Our analysis of handgun characteristics was extensive but not exhaustive. Our sampling technique did not include handgun models from small domestic manufacturers (63 pistol manufacturers and 11 revolver manufacturers), which account for approximately 8% of domestically produced pistols and 2% of domestically produced revolvers. On the basis of a review of the data in Gun Digest 28 on the models not included, there is no reason to expect that the characteristics of these handguns differed markedly from those of the handguns we examined.

It is important to note that our study was cross-sectional. We restricted our analysis to handgun models that were produced in 1996. The safety features and characteristics found on 1996 models might differ somewhat from those found in previous or subsequent years of the same model. Since 1996, some of the models we studied might have been discontinued, and new models might have been introduced. In the absence of significant market or legislative changes, there is no reason to expect that the prevalence of safety features has changed dramatically since 1996.

Our study suggests that the factoring criteria do have some effect on the presence of handgun safety devices. Requiring all domestically manufactured handguns to meet the factoring criteria might lead to a modest increase in the prevalence of handgun safety devices, safer handgun designs, and, ultimately, fewer handgun deaths. It might be possible, however, to move beyond the relatively weak factoring criteria to develop effective performance-based handgun safety standards, similar to the safety standards for automobiles, that could result in a significant decrease in firearm-related injuries.