Methods

During the last week of August 1993, the Sacramento Police Department conducted a gun exchange program in which participants received vouchers tickets to games played by the local National Basketball Association franchise. The program was widely promoted in the local media.

One hundred and twenty seven people participated. Police recorded a brief description of the gun; the participant's name, address, and date of birth were recorded if consent was given. No background checks or criminal investigations of participants were conducted. Eleven participants did not permit the police to record identifying information, leaving 116 potential respondents for our study. We designed the questionnaire and conducted the survey following a method proven to be successful in varied populations.^13,14 The survey was conducted from February through April 1994.

The questionnaire contained 50 questions and covered: (1) administration of the program itself, (2) reasons for turning in a gun, (3) experience with guns, (4) the guns exchanged, (5) current gun ownership, (6) storage of guns, and (7) demographics. Of these 50 questions, 19 came from five previously published national public opinion polls of gun owners (Gallup Organization, Gallup Poll, March 1991; Yankelovich Clancy Shulman for Time/CNN, 1989; Gallup Organization, Gallup Poll, May 1991; National Opinion Research Center, General Social Survey, 1991; Gallup Organization, Gallup Poll, November 1991). Questions taken from these polls were obtained by a computerized search of public opinion survey data from the Roper Center for Public Opinion Research at the University of Connecticut. These questions allowed us to compare gun owners who responded to our survey with gun owners nationwide. (Available resources did not permit a separate community-wide survey of gun ownership and usage practices in our study area.)

Participant identifying information and descriptions of the weapons exchanged, including type and caliber, were obtained from police records filled out at the time of the program. Weapon types included handguns (pistols, revolvers), long guns (rifles, shotguns), and airguns. Handguns were classified as small caliber (.22, .25, .32), medium caliber (.38, .380, 9 mm), or large caliber (.357, .45). An identification number linked questionnaires with the police department's records of the exchange program.

The odds ratios (ORs) were calculated using confidence interval (CI) analysis software, and exact tests were calculated using Epi-Info software, with ∝ = 0.05 chosen as the level of statistical significance.

Results

The response rate was 79% (92/116). Most respondents (62%) were male. Forty per cent were more than 55 years of age, 15% were less than 35, and none was less than 25. At the time of the program, 25% of respondents had a child under 14 years of age living with them; 13% had a young adult 15 to 24 years of age at home.

Seventy four respondents (80%) owned guns at the time of the program; the remaining 18 (20%) turned in guns that they had found or that were owned by others. Compared with gun owners nationwide, gun owners among our respondents were older and of higher socioeconomic status (table 1).

Respondent gun owners gave evidence of both similarities and differences with gun owners nationwide in their experience with guns (table 2). They were less likely to have grown up with guns or to have had any formal training with guns, but were equally likely to have used a gun defensively or to have known someone who had been shot in a violent crime. In addition, 32% of respondent gun owners knew someone who had used a gun to commit or attempt suicide; no comparison data were available.

Almost half the respondents (46%) cited concern that children might get and use the gun as an “important” reason for their participation in the program (fig 1). Fewer than 15%, however, reported a previous bad experience with a gun or concern for a possible shooting in their home as important. Comments illustrating the range of participants' reasons for exchanging guns are presented in table 3.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Percentage of respondents indicating that each of six possible reasons for turning in a gun was “important” in their decision to participate in the exchange program.

Of 141 firearms exchanged, 40 (28%) were long guns and 101 (72%) were handguns, including 66 revolvers and 35 pistols. Of the handguns, 63% were small caliber, 26% were medium caliber, 8% were large caliber, and 3% were of unknown caliber. Participants also exchanged 10 airguns, three Tasers (hand held devices that deliver electric shocks), and numerous boxes of ammunition. Most participants (87%) turned in one firearm, 8% turned in two or more, and 5% turned in only airguns. Almost one fourth of the respondents (23%) indicated the gun was not in working order, and another 24% were not sure. Most respondents (85%) had the gun in their household for more than one year.

Of respondents who were gun owners at the time of the program, 29 (41%) no longer owned any guns at the time of our survey 6–8 months later. The percentage of responding gun owners with more than one gun declined from 62% to 35%. The prevalence of handgun ownership among responding gun owners was reduced from 79% to 32%.

Forty three respondents retained at least one gun in their household after participating in the program. Of these, 23 (53%) still kept a handgun and 30 (70%) still kept a long gun (table 4). These continuing gun owners were just as likely as gun owners nationwide to keep a gun loaded, to store some or all of their guns unlocked, and to carry a gun on their person or in their car or truck. Continuing gun owners, however, were less likely than gun owners nationwide (OR 0.6, 95% CI, 0.3 to 1.0) to retain more than one gun in their household.

Discussion

Exposure to a gun—particularly a handgun—has repeatedly been associated with a substantial increase in risk of fatal firearm violence.^9–12 Our findings suggest that reduction in risk of firearm violence occurs for participants in gun exchange programs and members of their households, in that the prevalence of both gun ownership and handgun ownership have been reduced. Forty one per cent of gun owners no longer had any guns after the program, and the prevalence of handgun ownership declined by well over one half.

However, the community-wide benefit of such programs is probably less than such findings might suggest, for several reasons. First, while many participants no longer had any guns after the program, those who still had guns used them as other gun owners do, often keeping a gun loaded and unlocked, or concealed on their person or in their vehicle. These behaviors may be associated with a higher risk of firearm injury.

Second, as in other such programs,^7,8 participants in Sacramento's gun exchange program were older than other gun owners. No more than two of all 127 participants in the program were estimated to be less than 25years of age; neither provided identifier data and both were necessarily excluded from our survey (personal communication, A Kozuma, Sacramento Police Department). Females made up a higher percentage of participants than of the general population of gun owners.

Participants in these programs may therefore be at lower risk of firearm violence—or firearm homicide, at least—than are gun owners generally, even before exchanging any firearms. We note, however, that rates of firearm suicide are highest among older males.¹⁵ It may be that gun exchange programs have a greater potential to lower rates of firearm suicide than of firearm homicide. This remains a matter for further investigation; the one study to address this question found no such effect, but the number of events both before and after the program was too small for any conclusions to be drawn.⁷

In addition, a quarter of the guns exchanged in this program were reported not to be in working order and thus posed little potential for violence in the first place. This is a greater proportion than in the one other site from which data are available; in Seattle, 83% of firearms were reported to be operational and the status of another 14% was unknown.⁷

Finally, guns returned during exchange programs may not resemble the guns most frequently used in crime. For example, semiautomatic pistols made up 17% of buy-back guns in the city of Boston in 1993–94, but 32% of firearms confiscated by that city's police department from persons age 30 and older, and 52% of firearms confiscated from persons age 21 and younger, in that city during 1991–95. Long guns (rifles and shotguns) made up 42% of buy-back guns and a similar 39% of firearms confiscated by police from persons age 30 and older, but only 18% of firearms confiscated from persons age 21 and younger.¹⁶

Data from a separate study of firearms confiscated by police in Sacramento in 1995 suggest that similar differences exist in our study area.¹⁷ Long guns made up 28% of the exchanged guns in the present study but 18% of confiscated firearms. Semiautomatic pistols made up only 35% of the handguns in the exchange program and 57% of the handguns confiscated by police in 1995.

There are several limitations to this study. We did not have any information on the 11 program participants to whom questionnaires could not be sent or on the 24 non-respondents to the questionnaire. Since this was a relatively small survey of one city's gun exchange program, it may not be appropriate to generalize our results to other such programs, although results for other sites are similar.^7,8,16 Due to resource limitations we were not able to gather data from the general population of gun owners in Sacramento for comparison purposes, and it is possible that the observed differences between our respondents and gun owners nationwide are related to factors such as regional variations in gun usage rather than participation in the gun exchange program.

What are the implications of these findings for persons conducting, or considering, gun exchange programs? Concern that children might gain access to a gun was the most frequently cited reason for participating. Persons planning future programs might want to emphasize their potential benefits for children and young adults.

At the same time, the dearth of young adults among program participants poses a major challenge. The same problem has been noted elsewhere.^7,8 Our data cannot determine if this was due to a perceived need by young people to keep a gun, the type of incentive offered, the involvement of the police department, or ineffective advertising. For gun exchange programs to be optimally effective, future efforts must work to include high risk youth. As rates of violent crime among young people decrease, the perceived need to own a firearm for protection may decrease as well, making this task easier.

Lastly, our respondents who still owned guns often kept them loaded and easily accessible. Information on the risks that may be associated with those behaviors should be provided to program participants.

Communities may also wish to consider other measures to prevent firearm violence that rely at least in part on reducing the number of firearms available for criminal use. In a controlled experiment in Kansas City, law enforcement efforts focused on confiscating illegal firearms resulted in a substantial reduction in firearm crime.¹⁸ In Boston, tracing of all confiscated firearms has increasingly made it possible to disrupt the activities of firearms traffickers, which has contributed to a substantial decrease in homicide rates.¹⁹ In St Louis, a voluntary search and seizure program has resulted in the confiscation of several hundred firearms per year; an outcome evaluation is pending.²⁰ As these and other programs are replicated and evaluated, the relative cost effectiveness of gun exchange programs will be better understood.