Article Text

Association between firearm ownership, firearm-related risk and risk reduction behaviours and alcohol-related risk behaviours
  1. Garen J Wintemute
  1. Correspondence to Dr Garen J Wintemute, Violence Prevention Research Program, School of Medicine, University of California, Davis, 2315 Stockton Blvd, Sacramento, CA 95817, USA; gjwintemute{at}ucdavis.edu

Abstract

Alcohol use and firearm ownership are risk factors for violent injury and death. To determine whether firearm ownership and specific firearm-related behaviours are associated with alcohol-related risk behaviours, the author conducted a cross-sectional study using Behavioral Risk Factor Surveillance System data for eight states in the USA from 1996 to 1997 (the most recent data available). Altogether, 15 474 respondents provided information on firearm exposure. After adjustment for demographics and state of residence, firearm owners were more likely than those with no firearms at home to have ≥5 drinks on one occasion (OR 1.32; 95% CI 1.16 to 1.50), to drink and drive (OR 1.79; 95% CI 1.34 to 2.39) and to have ≥60 drinks per month (OR 1.45; 95% CI 1.14 to 1.83). Heavy alcohol use was most common among firearm owners who also engaged in behaviours such as carrying a firearm for protection against other people and keeping a firearm at home that was both loaded and not locked away. The author concludes that firearm ownership and specific firearm-related behaviours are associated with alcohol-related risk behaviours.

  • Firearms
  • alcohol
  • homicide
  • suicide
  • violence
  • aggression
  • alcohol/drugs

Statistics from Altmetric.com

Introduction

There are 260–300 million firearms in civilian hands in the USA.1 Firearms were kept in 35% of American households in 2006 and 22% of individuals reported that they owned firearms.2 In 2009 there were 31 228 firearm-related deaths in the USA3 and 66 769 persons had non-fatal gunshot wounds.4 Firearm ownership and household exposure to firearms are well-established risk factors for death from homicide and suicide.5–7

Alcohol use is likewise a risk factor for violent death.6–9 Approximately 34.5% of suicide and homicide victims in the USA tested positive for alcohol in 2007, of whom about 60% met legal criteria for acute intoxication.10

The possibility that firearm ownership and firearm-related risk behaviours are linked to heavy alcohol use and alcohol-related risk behaviours is therefore of particular interest. The existing evidence has important limitations. Studies have been restricted geographically or demographically,11–15 have involved small samples11–16 and have considered a limited range of firearm- and alcohol-related risk behaviours.11–17

The most comprehensive data on risk behaviours among adults in the USA come from the Behavioural Risk Factor Surveillance System (BRFSS) conducted under the auspices of the Centers for Disease Control and Prevention (CDC). We use BRFSS data to investigate associations between firearm ownership and alcohol use. Our data come from eight states and from 1996 to 1997; no more recent data exist for firearm ownership. We have two hypotheses: (1) alcohol-related risk behaviours are more common among firearm owners than among persons with no firearms at home; (2) for firearm owners, alcohol-related risk behaviours are more common among those who also engage in firearm-related risk behaviours.

Methods

The BRFSS survey

Details of survey design and administration and sample characteristics are available in the text and tables 1–3 of the online appendix.

Questions on firearms and alcohol (table 1 in online Appendix) were last fielded in 1997 by Colorado, Hawaii, Mississippi, New Hampshire, New Jersey, North Dakota and Ohio. Alaska fielded identical questions in 1996. (Kentucky and New York did as well, but only for half their survey samples and without explaining their selection criteria so were excluded).

The alcohol screening question asked whether, in the previous month, respondents had consumed ‘at least one drink of any alcoholic beverage such as beer, wine, wine coolers or liquor’. If so, follow-up questions addressed quantity, frequency and circumstance. CDC recoded responses to identify three risk behaviours, all occurring in the month prior to the survey: (1) having ≥5 drinks on one occasion; (2) driving after drinking ‘perhaps too much’; and (3) having ≥60 drinks per month. Non-drinkers and drinkers whose consumption fell below the criterion for a risk behaviour were both coded as not exhibiting that behaviour.

The firearm screening question asked whether there were ‘firearms now kept in or around [the respondent's] home … includ[ing] those kept in a garage, outdoor storage area, car, truck, or other motor vehicle’. If the response was yes, follow-up questions asked whether any of the firearms belonged to the respondent personally and whether any were at that time both loaded and unlocked. Respondents with firearms at home were asked whether they had carried a loaded firearm on their person in the previous month, outside the home and for protection against people; driven or travelled as a passenger in a vehicle in which they knew there was a loaded firearm, also in the previous month; and confronted another person with a firearm in the previous year. They were also asked about one potential risk reduction behaviour: attending a firearm safety workshop in the previous 3 years.

Questions on firearm-related behaviours were not asked of respondents with no firearms at home, of those who did not know whether firearms were present or of those who refused to answer the screening question.

Data analysis

Exposure to firearms was categorised as no firearms at home; household exposure (there are firearms at home, but the respondent is not a firearm owner); or personal firearm ownership. CDC grouped race/ethnicity into white, black, Hispanic and other (Native American, Asian and Pacific Islander).

Logistic regression was used throughout, with differences assessed using ORs and 95% CIs. Alcohol behaviours were the dependent variables in the main analysis. Ohio had the lowest prevalence of alcohol consumption and served as the referent state. All explanatory variables were significant in bivariate models and were retained in multivariate models. Respondents with household exposure to firearms were excluded from analyses involving firearm-related behaviours; there were few such persons and their prevalence of firearm-related behaviours was low.

Data analysis was performed using SAS Version 9.1.3 for Windows.

Results

Of 16 280 survey respondents, 15 474 (95.0%) provided specific information on their exposure to firearms. Of these, 3348 (21.6%) were firearm owners, 1800 (11.6%) non-owners lived in homes with firearms and 10 326 (66.7%) had no firearms at home (another 25 persons reported a firearm at home but did not give their exposure status).

The prevalence of firearms in the home and of each of the firearm-related behaviours varied with state of residence, was more common among men than women and among whites than others, and was least common among elderly persons (table 1). Personal firearm ownership ranged from 8.4% in Hawaii to 44.8% in Alaska and from 8.5% among women to 40.0% among men (figure 1 and online Appendix, table 4).

Table 1

Firearm exposure and firearm-related behaviours by state of residence and demographic characteristics

Figure 1

Firearm exposure by state.

Nearly all respondents (n=16 225, 99.7%) provided information on alcohol use. Of these, 8015 (49.4%) reported alcohol consumption in the previous month. The prevalence of alcohol use and of alcohol-related risk behaviours also varied with state of residence, was greater among men than among women and was inversely related to age (table 2).

Table 2

Alcohol use and alcohol-related risk behaviours by state of residence, demographic characteristics, firearms exposure and presence or absence of specific firearms-related behaviour

Main findings

Compared with respondents with no firearms at home (table 2), firearm owners were more likely to use alcohol (OR 1.66; 95% CI 1.54 to 1.80), to consume ≥5 drinks on one occasion (OR 2.01; 95% CI 1.81 to 2.23), to drink and drive (OR 2.42; 95% CI 1.90 to 3.09) and to consume ≥60 drinks per month (OR 2.39; 95% CI 1.95 to 2.93). No such increases were seen among non-owners with household exposure to firearms.

Firearm owners who engaged in firearm-related risk behaviours were in most cases at highest risk for alcohol-related risk behaviours (table 2). Firearm owners who did not engage in firearm-related risk behaviours were at intermediate risk. For example, for drinking and driving the risk was highest among firearm owners who drove or rode in a vehicle with a loaded firearm (OR 4.30; 95% CI 2.89 to 6.40) and lower (but still increased) for those who did not (OR 2.12; 95% CI 1.62 to 2.76). Conversely, firearm owners who attended a safety workshop were somewhat less likely than were those who did not to drink and drive or consume ≥60 drinks per month.

Multivariate models

In multivariate models (table 3) firearm owners were more likely than persons with no firearms at home to consume alcohol (OR 1.34; 95% CI 1.21 to 1.47), to consume ≥5 drinks on one occasion (OR 1.32; 95% CI 1.16 to 1.50), to drink and drive (OR 1.79; 95% CI 1.34 to 2.39) and to consume ≥60 drinks per month (OR 1.45; 95% CI 1.14 to 1.83). Firearm owners who engaged in firearm-related risk behaviours were again generally more likely than others to report alcohol-related risk behaviours, and were in all cases more likely to do so than were persons with no firearms at home. Firearm owners who attended a firearm safety workshop remained less likely to engage in alcohol-related risk behaviours than were those who did not.

Table 3

Multivariate analysis of alcohol use and alcohol-related risk behaviours by state of residence, demographic characteristics, exposure to firearms and among firearm owners by presence or absence of specific firearms-related behaviour*

Discussion

In this study population, firearm owners were more likely to report alcohol-related risk behaviours than were persons with no firearms at home and were generally at higher risk for these behaviours than were non-owners who lived in homes with firearms. Among firearm owners, those who engaged in firearm-related risk behaviours were in most cases more likely than others to engage in alcohol-related risk behaviours. Firearm owners who engaged in a firearm-related risk reduction measure were less likely than others to engage in alcohol-related risk behaviours.

Previous findings

These findings are in agreement with those of earlier and more limited studies.11–17 In perhaps the first such study, firearm owners were more likely to become ‘drunk several times a month or more’ than were non-owners.11 Persons who were planning to obtain a ‘concealed gun licence’ were more likely than others to drink frequently and/or heavily.12 Among young adult college students, firearm ownership was associated with several alcohol-related risk behaviours and adverse events.13 14 In Oregon, storing loaded and unlocked firearms at home was associated with heavy episodic and chronic alcohol consumption.15 In the 2001 National Gun Policy Survey, persons who carried firearms were more likely than others to ‘sometimes drink more than they should’.16 In the National Comorbidity Study replication, threatening others with a firearm was more common among persons with histories of alcohol abuse or alcohol dependence.17

There are numerous mechanisms by which risk behaviours involving alcohol and firearms might be linked or occur concurrently. Alcohol intoxication might impair judgement, for example, such that firearm-related risk behaviours are engaged in that would otherwise be avoided.18–20 This is of particular concern given that alcohol intoxication also impairs a firearm user's decision-to-shoot judgement and his accuracy.21 Alternatively, underlying and durable personality traits, such as impulsivity or a broad predisposition to risk taking, could lead to increases in both alcohol-related and firearm-related risk behaviours.18–20

Limitations of the study

These data are based on self-reports and may be affected by sampling error and response bias. The findings are for eight self-selected states and their generalisability is not known. Questions on firearm-related risk behaviours were not asked of those reporting no firearms at home. Carrying a firearm for protection or having a loaded and unlocked firearm at home would be rare or non-existent in that group, but riding in a vehicle with a loaded firearm might not.

The data were collected in the mid-1990s. The prevalence of firearm ownership has decreased from approximately 28% at that time to 22% in 20062 while the prevalence of alcohol use, including heavy use, has remained stable.22 It is possible, though unlikely, that the associations seen here have not persisted. A new exploration of the relationship between firearm ownership and alcohol use would be very helpful. States should incorporate the BRFSS firearms and alcohol modules in future surveys if those questions are not incorporated into the core survey. Additional research should address causation and estimate how often alcohol-related and firearm-related risk behaviours occur simultaneously.

Implications

New research is particularly important since legislation authorising the public carrying of loaded concealed firearms has become almost universal in the USA.23 In most states, carry permits are available to anyone who is not prohibited from owning firearms. Four states allow concealed firearms in bars provided that the armed person does not consume alcohol.24 Federal law does not restrict the purchase or possession of firearms by abusers of alcohol, and only a minority of states do so.25 Alcohol abusers are sometimes defined in statute simply as ‘habitual drunkards’ or ‘alcoholics’, making these statutes very difficult to enforce.26

From 1997, when BRFSS last collected data on firearms and alcohol, through 2009, an estimated 395 366 persons suffered firearm-related deaths.3 4 It is probable that more than a third of these deaths involved alcohol. Efforts to uncouple the use of firearms from the use of alcohol may have important benefits for the health and safety of the public.

What is already known on this subject

  • Firearm ownership may be associated with alcohol-related risk behaviours.

  • Available studies involve small samples, are limited demographically or geographically, and examine a small range of variables.

What this study adds

  • Firearm ownership itself and an array of firearm-related risk behaviours are associated with an array of alcohol-related risk behaviours.

  • Alcohol-related risk behaviours are most common among firearm owners who also engage in firearm-related risk behaviours.

  • Alcohol-related risk behaviours are less common among firearm owners who engage in a firearm-related risk reduction behaviour.

References

View Abstract

Supplementary materials

  • Web Only Data

    Files in this Data Supplement:

Footnotes

  • Funding This work was supported by grants from the Joyce Foundation and the California Wellness Foundation.

  • Competing interests None.

  • Ethics approval The institutional review board of the author's institution determined that this study was exempted from review.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.