Methods

TIMELINE

The first of three surveys was conducted from 2–18 May 1995. The second was conducted 18 months later, from 7–20 November 1996. The third was conducted two years later, from 18–22 January 1999.

SURVEY

All three surveys utilized random digit dialing to contact a stratified sample of 500 adults in the survey area. One hundred citizens in each of the five study counties were interviewed. Screening questions were asked to determine that the respondent was aged 21 years or older, and a member of the contacted household. Once these facts were established, our interviewers asked questions about household ownership, acquisition, storage, and disposition of firearms, as well as the respondent's rating of the effectiveness of a variety of home security measures. Participants were advised they could refuse to answer any question, and that they were free to terminate the interview at any time. The survey was approved by the Human Investigations Committee of Emory University.

DATA ANALYSIS

Responses were weighted on the basis of the population of each county to generate accurate aggregate estimates for the five county metro Atlanta area. Frequencies and cross tabulations were performed for each question based on demographic variables (age, race, sex, and level of education). Group comparisons were calculated using the χ² test, Yates' corrected χ² and Fisher's exact (for small cell size) for categorical variables, and the Mann-Whitney U test for ordinal values. For most responses, the statistical margin of error at the 95% confidence level was plus or minus four percentage points. The threshhold for statistical significance was set at p<0.05.

Results

DESCRIPTION OF THE POPULATION

Slightly less than two thirds of respondents to the 1995 survey were female. Sixty one per cent were white, and 34% were African-American. The mean age of respondents was 41 years, and 32% were college graduates. Two thirds of respondents owned their own home; the remaining one third rented. Thirty six per cent had one or more children aged 12 years or younger living in their home. The demographic breakdown of respondents in 1996 and 1999 resembled that of the 1995 group. With the exception of gender (60% of our respondents were female), the three groups closely resemble the demographic breakdown of adults in the study area (table 1). Among those who reported that guns are kept in their home, 29% (three year average) stated they had one or more children aged 12 years of age or younger living in the home; 15% (three year average) reported they had one or more teenagers aged 13–18 years living in the home, and 50% (three year average) had at least one child 18 years or younger living in the home.

FIREARM OWNERSHIP

Thirty eight per cent of respondents to the 1995 survey reported that one or more firearms are kept in their home. Seventy six per cent of firearm owning households had one or more handguns. In the 1996 survey, 41% of respondents reported a gun was kept in their household, and 72% of gun owning households had one or more handguns. In the 1999 survey, 35% of respondents reported that a gun was kept in their household and 69% of these gun owning households reported a handgun was kept in the home. None of these year-to-year differences are statistically significant. In all three surveys, white respondents were more likely than black respondents to keep a gun in their home (44% v 28%, three year average, p=0.000), but black gun owners were more likely to keep a handgun than whites (85% v 67%, three year average, p=0.0002).

FIREARM ACQUISITION, TRANSFER, AND DISPOSAL

In each year, gun owning respondents were much more likely to report acquiring a firearm than disposing of one. An average of 22% (24% in 1995, 19% in 1996, 22% in 1999) of gun owning households in each survey year reported buying a firearm within the past year. Not more than 7% of respondents reported selling, loaning, or giving away one or more firearms in the previous year in any of the three surveys. Less than 1% of gun owning respondents in any survey reported disposing of a firearm in the previous year through a gun “buy-back” program.

FIREARM STORAGE PRACTICES

Respondents in half of the gun owning households in the 1995 survey reported that they kept one or more guns in their home unlocked (table 2), and 44% kept at least one gun in their house loaded. Forty one per cent reported that they carried a gun in their car. In 1996, 47% of gun owning respondents kept one or more guns unlocked, 38% reported that a gun was kept loaded, and 40% reported that they carried a gun in their car. By 1999, the fraction of gun owning respondents who reported that they keep at least one gun unlocked had declined to 41% but the percentage of those who keep a gun loaded remained stable at 40%; 34% reported carrying a gun in their car. The decrease in persons leaving a gun unlocked between 1995 and 1998, and 1996 and 1998 is statistically significant (p = 0.05).

Two subgroups within the population of gun owners may have altered their firearm storage practices over the course of the three surveys. The percentage of male gun owners who stated they keep a firearm loaded in their home decreased from 55% in 1995 to 37% in 1996 (χ² 5.70, p=0.0170). In 1999, 43% of male gun owners kept a loaded firearm, this difference was not statistically significant from 1995 or 1996. The percentage who reported keeping a gun in their home unlocked also decreased, from 63% in 1995 to 45% in 1996 (χ² = 5.86, p=0.0155). This proportion remained stable in 1999 (χ² for 1995 v 1999, p = 0.015).

The percentage of African-American gun owners who reported that they keep a loaded firearm in their home declined from 54% in 1995 to 32% in 1996, to 49% in 1999. This trend was significant for 1995–96 (χ² = 4.47, p=0.0344) but not for 1996–99. African-American male gun owners in our group of respondents were also less likely to report keeping a gun unlocked (Yates' corrected χ² = 8.63, p=.0033; Fisher's exact p=0.0015) or loaded (Yates' corrected χ² = 4.58, p=0.0324; Fisher's exact p=0.0157) in 1996 than in 1995. No further decline was noted between the 1996 and the 1999 surveys.

REASONS FOR GUN OWNERSHIP

All three surveys asked gun owning respondents to identify the single most important reason that one or more guns were kept in their home. A majority of respondents to all three surveys (56% in 1995, 55% in 1996, and 53% in 1999) cited protection. Other important reasons for gun ownership included hunting or target practice (three survey average 23%); inherited the gun (three survey average 8%); collector's item (three survey average 6%); and occupational use (three survey average 4%). None of these responses varied by more than two percentage points in one survey compared with another.

African-Americans were more likely than whites to cite protection as their primary reason for gun ownership (three survey average 71% v 48%, p=0.0000). White respondents, on the other hand, were more likely to cite hunting (three survey average 27% v 11%, p=0.0002).

PERCEIVED BALANCE OF BENEFITS AND RISKS OF KEEPING A GUN IN THE HOME

To ascertain respondents' views of the balance of benefits and risks of keeping a gun in the home, they were read three statements about the balance of benefits and risks of keeping a gun in the house and asked to rate their level of agreement with each statement on an 11 point scale (0 = strongly disagree to 10 = strongly agree).

For purposes of analysis, responses to each statement were aggregated into three groups: 0–3 = strongly disagree, 4–6 = somewhat agree/disagree, and 7–10 = strongly agree. In each survey year, a plurality of respondents (44.1%) strongly disagreed with the statement “A home with a gun is more secure than a home without a gun, because a gun can be used for protection”. Clear majorities in each year (55%) strongly agreed with the second statement “A home with a gun is less secure than a home without a gun, because a gun can be involved in an accidental shooting, suicide, or family homicide” compared with 20% (three year average) that strongly disagreed. A larger plurality disagreed (three year average 43.8%) than agreed (three year average 30.7%) with the final statement “On balance, a home with a gun is safer than the same home without a gun”. Responses were generally consistent across all three surveys, reflecting little shift in public opinion (table 3).

Not surprisingly, gun owners rated the protective value of keeping a gun in the home more favorably than non-gun owners. Owners also expressed less concern about the associated risks of keeping a gun in the home. An average of 45% (47% of gun owners in 1995, 46% in 1996 and 43% in 1999) “strongly agreed” with the statement “On balance, a home with a gun is safer than the same home without a gun.” In contrast, only 20% of non-gun owners (21% in 1995, 18% in 1996, and 22% in 1999) strongly agreed with this statement. These differences were highly significant (p = 0.001 level) during every year of the survey (table 3).

EFFECTIVENESS OF HOME SECURITY MEASURES

In the first two editions of the survey, respondents were asked to rate the effectiveness of five home security measures on an 11 point scale ranging from zero (“not at all effective”) to 10 (“very effective”). “Installing a burglar alarm system” was rated most effective of the five options (mean scores 7.76 in 1995, and 7.90 in 1996). “Having a gun in the home” was rated least effective in 1995 and 1996 (mean scores 4.46 and 4.41, p<0.001).

“Installing bright exterior lights”, “having a dog in the home”, and “installing metal security doors” received mean scores of between 6.28 and 7.50 in either year (7.41, 6.69, and 6.28 respectively in 1995, and mean scores of 7.50, 6.77, and 6.28 in 1996) (table 4).

As expected, gun owners consistently rated the effectiveness of “having a gun in the home” more favorably than non-gun owners (mean scores 6.52 v 3.11, Mann-Whitney U p<0.001 in 1995; mean score 6.27 v 2.98, Mann-Whitney U p<0.001 in 1996). Despite this fact, gun owning respondents rated having a gun in the home less effective than installing a burglar alarm system, bright exterior lights or having a dog in the home. Non-gun owners rated having a gun in the home last among the five options in both the 1995 and 1996 surveys (table 4).

Discussion

The rate of gun ownership reported by respondents to our surveys is lower than the national average, but fairly typical of urban residents.^8,9 Although it is possible that some respondents gave false negative replies to the questions about ownership of a firearm, we consider this unlikely. To encourage accurate replies, questions about gun ownership were preceded by the permissive statement “Half of all homes contain one or more firearms; are guns of any kind kept in your home?” We did not ask respondents if any of the guns were unregistered or illegally owned. Respondents were informed that they could refuse to answer any question or terminate the interview at any time. Previous research has demonstrated that respondent answers to survey questions about gun ownership are generally valid.^10,11

A majority of gun owning respondents cited protection as their most important reason for keeping a gun. Not surprisingly, they also viewed the protective benefits of gun ownership more favorably than non-gun owners. However, a growing body of research suggests that keeping a loaded and readily available gun in the home does more harm than good.^12–20

Our findings suggest many gun owning households may be ambivalent about their decision to keep a gun in the home. Over the three surveys, an average of 37% of gun owning respondents “strongly agreed” with the statement, “A home with a gun is less secure than a home without a gun, because a gun can be involved in an accidental shooting, suicide or family homicide”. A minority of gun owners (three year average 45%) “strongly agreed” with the statement, “On balance, a home with a gun is safer than the same home without a gun”. The remaining 55% of gun owners were either neutral or disagreed with this statement. Non-gun owners were much more skeptical of the perceived the balance of benefits and risks associated with keeping a gun in the home.

There seems to be a basic incongruence between practices and the beliefs of many adults who keep a firearm in their home. Although a majority of gun owners identified “protection” as their most important reason for keeping a gun in the house, most recognized that there are also dangers associated with doing so. It is interesting to note that the majority of respondents in gun owning as well as non-gun owning households considered several non-lethal home security measures to be more effective than keeping a gun in the home. We are unaware of any empirical data that supports or refutes this belief.

Our findings are limited in certain respects. They are based on three small (n=500) surveys of adults living in a single metropolitan area and therefore may not be generalizable to other cities in the US. However, a larger telephone survey of a nationally representative sample of 4138 registered voters in urban areas of the US yielded results that are similar to our own. In that survey, 29% of respondents reported that they believed keeping a gun in the home makes the home more safe, 40% believed it made the home less safe, 23% said it depends, and 9% were not sure.²¹

Sixty per cent of our respondents were women; most gun owners are men.⁹ Ludwig and colleagues have noted that wives tend to report household ownership of firearms less often than husbands.²² In all three of our surveys, women tended to view the merits of keeping a gun in the home less favorably than men (table 5). Others have made the same observation.²¹ If this is true, the preponderance of women respondents may bias our surveys in a direction that is less favorable to gun ownership.

Health providers interested in reducing the risk of firearms injuries in the home may find it easier to promote “safe storage” of guns than to counsel gun owning patients to remove firearms from their home.³ The National Rifle Association's A Parent's Guide to Gun Safety advises owners to “Always keep the gun unloaded until ready to use” and “store guns so that they are inaccessible to children and other unauthorized users”.²³ The Sporting Arms and Ammunition Manufacturers Institute supports the notion that the safe handling of firearms requires that guns be stored unloaded in a locked area separate from the ammunition.^4,24 The Clinician's Handbook of Preventive Services includes the following recommendations: (1) never keep a loaded gun in the home; (2) keep guns and ammunition locked in separate locations; (3) always treat a gun as if it were loaded; and (4) never allow children access to guns.²⁵ The American Medical Association has produced a guidebook for physicians that emphasizes the importance of safe storage.²⁶

Although no study has conclusively demonstrated that storing guns in a secure manner reduces the subsequent risk of injury, there is indirect evidence that safe storage of firearms is beneficial. Safe storage could reduce the risk that firearms will be lost due to burglary or theft, an event that occurs more than 350 000 times each year in the US.²⁷ Safe storage may also decrease the likelihood that a loaded gun will be reached by a child²⁸ or used to commit suicide.¹⁴

Implications for prevention

Firearm injuries are a major cause of premature death and disability nationwide.^29,30 In metropolitan Atlanta, most gun owning households contain a handgun. Many households keep at least one gun unlocked or loaded. Our findings suggest that a majority of these gun owning households are ambivalent about the balance of benefits and risks associated with keeping a gun in the home. Perhaps in response to these concerns, the number of gun owning households that report unsafe storage practices has declined somewhat in recent years. Health educators and health care providers should view this development as an encouraging trend and redouble their efforts to educate patients about the importance of safe storage^13,26,31 and the availability of non-lethal alternatives for home security.