Abstract
Objectives. To characterize firearm ownership and storage practices in Washington State and assess their relationship with suicide risk factors.
Methods. Using Washington State Behavioral Risk Factor Surveillance System data for 2013 to 2016, we conducted survey-weighted multivariable Poisson regression models to obtain prevalence ratios (PRs) and confidence intervals (CIs).
Results. Of 34 884 adult respondents, 34.3% (95% CI = 33.7%, 35.0%) reported a firearm in their household, among whom 36.6% (95% CI = 35.4%, 37.7%) stored their firearm locked and unloaded. There were no differences in mental health indicators by firearm ownership or storage practice status. Binge and chronic alcohol use were somewhat more prevalent among adults from firearm-owning households (PR = 1.2; 95% CI = 1.1, 1.3; PR = 1.2; 95% CI = 1.1, 1.4, respectively) and among those living in households not practicing safe storage (PR = 1.4; 95% CI = 1.2, 1.7; PR = 1.9; 1.5, 2.3, respectively).
Conclusions. Variability in mental health does not explain the substantial increased suicide risk among individuals in firearm-owning households. Greater prevalence of alcohol misuse among adults in firearm-owning households not practicing safe storage highlights the need for suicide prevention interventions.
In the United States, suicide is a leading cause of death, with nearly half of those deaths involving a firearm.1 Several ecological studies have found a relationship between population-level firearm ownership and suicide rates.2–4 Using varying levels of aggregation in defining geographic regions, these studies have found an association at state and US census region levels. At the individual level, multiple case–control studies have identified strong associations between home firearm ownership and suicide for all members of the household.5,6
Some have raised the possibility that the increased suicide risk among individuals from firearm-owning households may be mainly attributable to their differential baseline suicide risk, rather than the presence of a firearm in the home.7 Numerous studies have indirectly addressed this concern by examining potential differences in suicide risk factors between homes with and those without firearms.8–14 Several of those studies have found no significant differences in mental illness or substance use between members of firearm-owning versus nonowning households,8–11 although a few studies have observed some differences.12–14 Methodological work has shown that any such observed differences are insufficient to fully explain the strong association (i.e., an overall 300%–400% increased risk) between living in a firearm-owning household and suicide.15 Some of these studies focused explicitly on adults with children in the home.12 Other investigations used data from the General Social Survey, National Comorbidity Survey Replication, and Primary Care Research in Substance Abuse and Mental Health for the Elderly study from the early 1990s to the mid-2000s, all of which found no relationship between mental health and firearm ownership.9,11,14
Among firearm-owning households, unsafe storage practices (keeping firearms unlocked or loaded) have been associated with a greater risk of firearm suicide among both youths and adults.5,16,17 Less is known about the relationship between safe storage practice and suicide risk factors. A few studies observed that home firearm access and unsafe storage practices were similarly common among households with youths who had and those who did not have mental illness,10,18 suggesting that efforts are not being taken to limit these young people’s access to lethal means. Although 1 study found a negative association between safe storage laws and rates of firearm suicide at the state level,19 our knowledge of the relationship between firearm storage practices and mental illness among adults is severely limited. Suicide is known to be an impulsive decision, with 1 study finding that nearly half of suicide attempt survivors reported less than 10 minutes between the first occurrence of a thought of suicide and their subsequent attempt.20 Although adults likely have the ability to unlock or load household firearms, delaying an individual’s access to lethal means may be beneficial in preventing suicide.21
To our knowledge, the most recent Behavioral Risk Factor Surveillance System (BRFSS) data used to examine the relationship between firearm ownership and suicide risk factors date to 2004,22,23 and the relationship between storage practice and suicide risk factors has not been previously studied using these data. Using data from BRFSS in Washington State, we conducted an investigation to (1) characterize adult residents by their household firearm ownership and storage practice status, (2) provide an updated analysis of the association of household firearm ownership with suicide risk factors, and (3) examine associations between household firearm storage practices and suicide risk factors.
On the basis of previous studies, we postulated that the scope and nature of firearm ownership can meaningfully vary by factors such as urbanicity, veteran status, and the presence of children in a home. Because these factors themselves are potentially associated with the risk of suicide, we sought to examine the prevalence of suicide risk factors by firearm ownership status across levels of these variables. Previous studies have also indicated that firearm ownership and the risk of suicide can vary by sociodemographic characteristics; male gender, middle age, unemployment, and living alone are all associated with increased rates of suicide.2,24–26 Our overarching goal was to add contemporary information to the literature about the association of firearm ownership and storage practices with suicide risk factors. This information can in turn further inform the policy and practice pertaining to suicide prevention.
METHODS
We used data from the 2013, 2015, and 2016 Washington State BRFSS surveys to address our goals. BRFSS uses random digit dialing to survey noninstitutionalized adults aged 18 years or older about health-related risk behaviors, chronic health conditions, and the use of preventative services. Selected information on alcohol consumption and mental health is collected nationwide. Washington also uses state-added sections to acquire additional information about mental health and household firearm ownership and storage practices. BRFSS data are made available by the Washington State Department of Health, Center for Health Statistics, Behavioral Risk Factor Surveillance System.
Firearm Ownership and Storage Practices
Firearm ownership and storage practices were assessed through a 4-question module. Survey participants were first asked: “Are any firearms now kept in or around your home?” Those reporting the presence of a firearm were further asked about storage practices. The follow-up questions asked: “Is there a firearm in or around your home that is now loaded?” and “Is there a firearm in or around your home that is now unlocked? By unlocked, we mean that you do not need a key or combination to get the firearm or to fire it. We don’t count a safety as a lock.” If individuals reported that there was at least 1 loaded firearm, they were subsequently asked: “Are any of the loaded firearms also unlocked? By unlocked, we mean you do not need a key or combination to get the firearm or to fire it. We don’t count a safety as a lock.”
We characterized storage practices among residents of firearm-owning households in 1 of 3 ways. Those who used the safest storage method were individuals who reported all household firearms being stored both locked and unloaded. An intermediate risk group included individuals reporting the presence of at least 1 household firearm that either was unlocked and unloaded or was locked and loaded. Individuals reporting that there was at least 1 firearm in or around the home that was both unlocked and loaded had the least safe storage practices and constituted the third category. BRFSS questions are ordered to distinguish an individual reporting living in a home with at least 1 unlocked firearm and at least 1 loaded firearm from an individual living in a home with a firearm stored unlocked and loaded. We excluded individuals not reporting storage methods from this further analysis.
Suicide Risk Factors
We included measures of mental health and substance use known to be associated with suicide. We defined depression diagnosis as reporting having ever being told by a doctor, nurse, or other health professional that the respondent had a depressive disorder, including depression, major depression, dysthymia, or minor depression. The Kessler 6 (K6) instrument was used to assess mental health, with serious mental illness being defined as a score of 13 or greater, as previously recommended.27
We considered individuals to have frequent mental distress if they reported 14 or more days of poor mental health in the past month.28 We additionally included measures of both chronic alcohol use and binge drinking. We defined chronic or excessive drinking as 8 or more servings of alcohol in 1 week for women and 15 or more servings in 1 week for men.29 We defined binge drinking as the consumption of 4 or more servings of alcohol at 1 time for women and 5 or more servings for men.29 Appendix A (available as a supplement to the online version of this article at http://www.ajph.org) provides the survey questions used to collect information on all measures.
Statistical Analysis
We used multivariable Poisson regression models to obtain prevalence ratios (PRs) and their 95% confidence intervals (CIs), comparing prevalence of suicide risk factors among individuals from firearm-owning households with that of individuals from households without firearms. Among persons in firearm-owning households, we compared suicide risk factors on the basis of household firearm storage practices. We selected model covariates a priori and included respondent age, race/ethnicity, gender, household income, education level, employment, and marital status as potential confounders. We identified urbanicity, veteran status, and the presence of children aged 18 years or younger residing in the home a priori as potential effect measure modifiers and assessed them using interaction terms in the regression models examining the relation between firearm ownership and suicide risk factors after controlling for potential confounders.
We weighted data using the iterative proportional fitting, or raking, methodology introduced to BRFSS in 2011, which seeks to minimize bias and make estimates more generalizable to the state population and to accommodate the growing use of cell phones and the decline in landline usage. Data collected before and those collected after the 2011 change are not directly comparable, and the Centers for Disease Control and Prevention advises against using both in the same analysis. The 2013, 2015, and 2016 surveys were the only years after this change when Washington added the firearms module to the survey.
Data were missing for key variables among some respondents. Many individuals without an answer to these questions were also missing answers to the previous module, leading us to believe that most of the missing data were attributable to survey fatigue. Assuming missingness at random, we conducted analyses of data imputed through chained equations. We conducted the analyses with Stata release 14.2 (StataCorp LP, College Station, TX).
RESULTS
During the 2013, 2015, and 2016 surveys, there were 41 537 respondents. We included 34 884 (84.0%) respondents in our analytic database after excluding records of individuals who refused to answer firearm module questions (n = 1664), stated they did not know whether there was a firearm in their household (n = 155), or did not complete the entire survey and therefore did not provide responses to the firearm module (n = 4834). Individuals not providing information on household firearm ownership were more likely to be male, be younger, identify as Asian, and have an annual household income of less than $75 000. The excluded individuals were less likely to identify as White and to be retired.
Overall, 34.3% (95% CI = 33.7%, 35.0%) of adults reported living in a firearm-owning household. Respondents from firearm-owning households were more likely to be older, White, male, retired, and married than were those from firearm nonowning households (Table 1). Respondents from firearm-owning households were also more likely to have reported an annual household income of greater than $50 000 and residence in a rural zip code. Neither college attendance (some college or completion) nor the presence of children in the household meaningfully varied by household firearm ownership, although individuals from firearm-owning households were more likely to be veterans (Table A, Appendix B, available as supplements to the online version of this article at http://www.ajph.org).
TABLE 1—
Characteristics of Washington State Adults by Household Firearm Ownership and Storage Practices: Behavioral Risk Factor Surveillance System, 2013, 2015, 2016
| Firearm-Owning Households |
|||||
| Characteristic | Non–Firearm-Owning Households (n = 22 177), Weighted % (95% CI) | Locked and Unloaded (n = 4361), Weighted % (95% CI) | Locked and Loaded/Unlocked and Unloaded (n = 5580), Weighted % (95% CI) | Unlocked and Loaded (n = 2402), Weighted % (95% CI) | Total (n = 34 884), Weighted % (95% CI) |
| All respondents | 65.7 (65.0, 66.3) | 12.3 (11.9, 13.0) | 15.0 (14.6, 16.0) | 6.3 (6.0, 7.0) | 100.0 |
| Age, y | |||||
| 18–24 | 12.3 (11.5, 13.0) | 12.3 (10.6, 14.0) | 9.2 (7.8, 10.5) | 8.2 (6.3, 10.2) | 11.6 (11.0, 12.2) |
| 25–44 | 35.1 (34.2, 36.0) | 31.8 (29.8, 33.8) | 29.2 (27.4, 31.0) | 23.7 (21.1, 26.5) | 33.0 (32.3, 33.7) |
| 45–64 | 34.0 (33.2, 34.9) | 36.4 (34.5, 38.2) | 38.3 (36.6, 40.0) | 41.4 (38.7, 44.1) | 35.4 (34.8, 36.1) |
| ≥ 65 | 18.6 (18.0, 19.1) | 19.6 (18.4, 20.9) | 23.4 (22.2, 24.6) | 26.7 (24.7, 28.8) | 19.9 (19.5, 20.4) |
| Race/ethnicity | |||||
| White | 68.2 (67.2, 69.1) | 85.2 (83.5, 86.9) | 87.5 (86.2, 88.9) | 88.2 (86.1, 90.2) | 74.5 (73.8, 75.2) |
| Black | 3.9 (3.5, 4.3) | 1.3 (0.7, 1.9) | 1.6 (1.0, 2.1) | 1.1 (0.4, 1.7) | 3.0 (2.7, 3.3) |
| Asian | 9.8 (9.0, 10.5) | 4.4 (3.3, 5.6) | 2.0 (1.3, 2.6) | 2.4 (1.1, 3.7) | 7.5 (6.9, 8.0) |
| Hispanic | 13.2 (12.5, 13.9) | 4.4 (3.4, 5.3) | 4.0 (3.1, 4.9) | 3.5 (2.4, 4.6) | 10.1 (9.5, 10.6) |
| Othera | 5.0 (4.6, 5.4) | 4.7 (3.9, 5.7) | 4.9 (4.2, 5.8) | 4.9 (3.9, 6.2) | 4.9 (4.6, 5.3) |
| Female | 55.7 (54.8, 56.6) | 51.5 (49.5, 53.5) | 39.9 (38.2, 41.6) | 33.2 (30.6, 35.7) | 51.6 (50.8, 52.3) |
| Income, $ | |||||
| < 25 000 | 27.6 (26.7, 28.5) | 14.3 (12.8, 16.0) | 12.0 (10.8, 13.3) | 14.3 (12.4, 16.4) | 22.6 (21.9, 23.3) |
| 25 000 < 50 000 | 24.3 (23.4, 25.2) | 21.2 (19.6, 23.0) | 23.1 (21.5, 24.7) | 25.2 (22.8, 27.8) | 23.8 (23.1, 24.5) |
| 50 000 < 75 000 | 14.3 (13.6, 14.9) | 19.5 (17.9, 21.2) | 19.1 (17.7, 20.6) | 21.3 (19.1, 23.8) | 16.1 (15.6, 16.7) |
| ≥ 75 000 | 33.9 (33.0, 34.8) | 45.0 (42.9, 47.1) | 45.8 (43.9, 47.7) | 39.2 (36.4, 42.1) | 37.5 (36.8, 38.2) |
| Education | |||||
| < high school | 13.4 (12.6, 14.2) | 6.3 (5.1, 7.8) | 6.2 (5.1, 7.5) | 6.0 (4.6, 7.7) | 10.9 (10.3, 11.5) |
| High school or GED | 22.5 (21.8, 23.3) | 28.0 (26.2, 29.9) | 26.4 (24.8, 28.0) | 31.2 (28.7, 33.9) | 24.4 (23.7, 25.0) |
| Some college | 33.4 (32.5, 34.3) | 37.6 (35.6, 39.6) | 39.8 (38.0, 41.6) | 42.5 (39.7, 45.3) | 35.5 (34.8, 36.2) |
| College graduate | 30.7 (30.0, 34.5) | 28.1 (26.6, 29.7) | 27.7 (26.3, 29.1) | 20.3 (18.5, 22.2) | 29.3 (28.7, 29.9) |
| Employment | |||||
| Employed | 56.1 (55.1, 57.0) | 56.2 (54.2, 58.2) | 60.1 (58.4, 61.7) | 55.2 (52.5, 57.9) | 56.6 (55.9, 57.3) |
| Elective unemploymentb | 30.6 (29.8, 31.4) | 33.8 (32.0, 35.6) | 31.7 (30.2, 33.3) | 33.4 (31.0, 35.8) | 31.3 (30.7, 32.0) |
| Unemployed | 13.4 (12.7, 14.0) | 10.0 (8.8, 11.4) | 8.2 (7.3, 9.3) | 11.4 (9.7, 13.4) | 12.0 (11.6, 12.6) |
| Marital status | |||||
| Married | 61.8 (60.9, 62.7) | 73.6 (71.6, 75.5) | 75.6 (73.9, 77.2) | 72.0 (69.3, 74.6) | 66.0 (65.2, 66.7) |
| Divorced/separated | 30.0 (29.2, 30.9) | 21.0 (19.2, 22.9) | 18.5 (17.1, 20.1) | 22.0 (19.6, 24.6) | 26.7 (26.0, 27.4) |
| Never married | 4.6 (4.2, 5.1) | 3.4 (2.6, 4.3) | 3.6 (2.9, 4.4) | 3.8 (2.7, 5.2) | 4.2 (3.9, 4.6) |
| Otherc | 3.6 (3.2, 4.0) | 2.1 (1.5, 2.8) | 2.4 (1.8, 3.1) | 2.2 (1.6, 3.2) | 3.1 (2.9, 3.4) |
Note. CI = confidence interval; GED = general equivalency diploma. For full breakdown, see Appendix B (available as a supplement to the online version of this article at http://www.ajph.org). Table does not present information on 367 respondents in firearm-owning households that were missing information on storage practices.
Includes Native Hawaiian/other Pacific Islander, American Indian/Alaska Native, multiracial, and other racial identities.
Includes self-reported students, homemakers, or retirees.
Includes individuals in registered domestic partnerships and those who are widowed.
We further characterized the 12 707 participants in firearm-owning households by household firearm storage practices (Table 1). The safest storage practice was reported by 36.6% (95% CI = 35.4%, 37.7%) of respondents from firearm-owning households; 18.8% (95% CI = 17.9%, 19.7%) of respondents from firearm-owning households reported the presence of at least 1 unlocked and loaded firearm (Table A; Appendix B). Individuals reporting that household firearms were both locked and unloaded tended to be younger, female, and college graduates and to report children younger than 18 years residing in the household than were individuals from households with intermediate- or high-risk storage practices. Respondents reporting safe storage practices were also less likely to be retired or a veteran or to report residing in a rural zip code than were individuals reporting unsafe practices.
The prevalence of mental health indicators did not differ significantly by household firearm ownership status (Table 2). Binge alcohol consumption (PR = 1.19; 95% CI = 1.10, 1.28), chronic alcohol consumption (PR = 1.21; 95% CI = 1.07, 1.37), and chronic pain (PR = 1.14; 95% CI = 1.09, 1.19) were more common among individuals in firearm-owning households. These findings were not meaningfully different in analyses of imputed data (Appendix C, available as a supplement to the online version of this article at http://www.ajph.org). The interaction between firearm ownership and urbanicity in relation to depression was statistically significant; depression diagnosis was less common among individuals in urban firearm-owning households (PR = 0.93; 95% CI = 0.87, 0.99) but did not vary by household firearm ownership in rural areas (PR = 1.09; 95% CI = 0.95, 1.26). There was no significant interaction between firearm ownership and urbanicity in association with any of the other risk factors. Veteran status did not modify the association between firearm ownership and any of the suicide risk factors.
TABLE 2—
Prevalence of Suicide Risk Factors by Household Firearm Ownership Among Washington State Adults: Behavioral Risk Factor Surveillance System, 2013, 2015, 2016
| Variable | No Firearms in Household, (n = 22 177), % (95% CI) | Firearm in Household, (n = 12 707), % (95% CI) | PR (95% CI) |
|
| Crude | Adjusteda | |||
| Depression diagnosis | 24.0 (23.2, 24.8) | 20.9 (19.9, 21.8) | 0.87 (0.82, 0.92) | 0.95 (0.89, 1.01) |
| Felt worthless (all or most of the time) | 2.7 (2.4, 3.0) | 1.9 (1.6, 2.3) | 0.71 (0.57, 0.88) | 0.99 (0.76, 1.30) |
| Felt everything was an effort (all or most of the time) | 7.4 (6.9, 8.0) | 5.6 (5.1, 6.2) | 0.76 (0.67, 0.86) | 1.01 (0.88, 1.16) |
| Serious mental illness (K6 of ≥ 13) | 4.0 (3.6, 4.4) | 2.8 (2.4, 3.3) | 0.72 (0.60, 0.86) | 1.03 (0.83, 1.27) |
| Frequent mental distress (≥ 14 d poor mental health) | 11.9 (11.3, 12.5) | 10.3 (9.6, 11.1) | 0.87 (0.79, 0.95) | 1.07 (0.96, 1.18) |
| Binge alcohol use | 14.6 (13.9, 15.3) | 18.7 (17.7, 19.7) | 1.28 (1.19, 1.38) | 1.19 (1.10, 1.28) |
| Chronic alcohol use | 5.9 (5.5, 6.3) | 8.0 (7.4, 8.7) | 1.35 (1.22, 1.51) | 1.21 (1.07, 1.37) |
| Chronic pain | 24.1 (23.4, 24.8) | 29.6 (28.6, 30.6) | 1.23 (1.18, 1.29) | 1.14 (1.09, 1.19) |
Note. CI = confidence interval; K6 = Kessler 6; PR = prevalence ratio.
Adjusted for age, race/ethnicity, gender, income, education, employment, marital status, and urbanicity.
There was evidence of statistical interaction between the presence of a child younger than 18 years in the home and household firearm ownership in relation to suicide risk factors. The interaction was statistically significant for previous depression diagnosis, having felt worthless most or all the time, serious mental illness as indicated by K6 score, frequent mental distress, and chronic pain. Despite statistical significance, the only indicator with evidence of meaningful effect modification was serious mental illness; in households with children, there was a positive association between household firearm ownership and serious mental illness (PR = 1.71; 95% CI = 1.17, 2.49), whereas in households without children, respondents from firearm-owning households had a lower prevalence of serious mental illness (PR = 0.78; 95% CI = 0.61, 0.99).
Among firearm-owning households, the prevalence of suicide risk factors varied by storage practice (Table 3). Compared with individuals from households storing firearms locked and unloaded, binge alcohol use was more common among individuals in homes with at least 1 firearm stored locked and loaded or unlocked and unloaded (PR = 1.17; 95% CI = 1.03, 1.33) and even more common among individuals in homes with at least 1 unlocked and loaded firearm (PR = 1.44; 95% CI = 1.23, 1.67). Chronic alcohol use was also more common among individuals in homes with at least 1 unlocked and loaded firearm (PR = 1.85; 95% CI = 1.46, 2.33) as was frequent mental distress (PR = 1.25; 95% CI = 1.01, 1.56).
TABLE 3—
Prevalence of Suicide Risk Factors by Household Firearm Storage Practices Among Washington State Adults Residing in Firearm-Owning Homes: Behavioral Risk Factor Surveillance System, 2013, 2015, 2016
| Variable | Locked and Unloaded (n = 4361), % (95% CI) | Locked and Loaded/Unlocked and Unloaded (n = 5580) |
Unlocked and Loaded (n = 2402) |
||||
| % (95% CI) | Crude PR (95% CI) | Adjusteda PR (95% CI) | % (95% CI) | Crude PR (95% CI) | Adjusteda PR (95% CI) | ||
| Depression diagnosis | 20.2 (18.6, 21.8) | 21.4 (19.9, 22.8) | 1.06 (0.95, 1.18) | 1.13 (1.01, 1.26) | 19.8 (17.6, 21.9) | 0.98 (0.86, 1.12) | 1.04 (0.90, 1.19) |
| Felt worthless (all or most of the time) | 1.8 (1.2, 2.4) | 1.6 (1.1, 2.1) | 0.91 (0.58, 1.42) | 1.10 (0.68, 1.78) | 2.4 (1.5, 3.2) | 1.30 (0.81, 2.10) | 1.29 (0.75, 2.22) |
| Felt everything was an effort (all or most of the time) | 5.1 (4.2, 6.1) | 5.6 (4.7, 6.5) | 1.09 (0.86, 1.39) | 1.18 (0.91, 1.53) | 6.6 (5.3, 8.0) | 1.29 (0.98, 1.70) | 1.27 (0.94, 1.71) |
| Serious mental illness (K6 of ≥ 13) | 3.0 (2.2, 3.8) | 2.6 (2.0, 3.3) | 0.87 (0.61, 1.25) | 1.21 (0.82, 1.78) | 2.6 (1.8, 3.5) | 0.87 (0.58, 1.33) | 0.89 (0.55, 1.43) |
| Frequent mental distress (≥ 14 d poor mental health) | 10.4 (9.1, 11.8) | 9.7 (8.6, 10.8) | 0.93 (0.78, 1.11) | 1.11 (0.92, 1.33) | 11.3 (9.5, 13.0) | 1.08 (0.88, 1.33) | 1.25 (1.01, 1.56) |
| Binge alcohol use | 15.9 (14.3, 17.4) | 19.3 (17.8, 20.8) | 1.22 (1.07, 1.38) | 1.17 (1.03, 1.33) | 23.5 (21.0, 26.0) | 1.48 (1.28, 1.71) | 1.44 (1.23, 1.67) |
| Chronic alcohol use | 6.5 (5.5, 7.5) | 7.9 (6.9, 8.8) | 1.22 (1.00, 1.48) | 1.21 (0.99, 1.48) | 11.8 (9.9, 13.6) | 1.82 (1.46, 2.27) | 1.85 (1.46, 2.33) |
| Chronic pain | 27.9 (26.2, 29.6) | 29.2 (27.7, 30.7) | 1.04 (0.96, 1.13) | 1.00 (0.94, 1.13) | 33.7 (31.2, 36.1) | 1.20 (1.09, 1.32) | 1.03 (0.94, 1.13) |
Note. CI = confidence interval; K6 = Kessler 6; PR = prevalence ratio. Locked and unloaded storage is the referent category for all PRs.
Adjusted for age, race/ethnicity, gender, income, education, employment, marital status, and urbanicity.
DISCUSSION
We believe this is the first study of firearm ownership and storage practices among a representative sample of adults on the basis of BRFSS data in more than a decade. In this study of Washington adults, the prevalence of mental health indicators was not markedly different between individuals from firearm-owning and those from nonowning households. Adults from firearm-owning households reported depression diagnosis, feeling worthless, feeling day-to-day tasks were an effort, serious mental illness, and frequent mental distress at a similar frequency as those from nonowning households.
Congruent with some previous studies,13,30 we found that the prevalence of alcohol misuse did vary somewhat by household firearm ownership, with binge and chronic drinking being more common among residents of firearm-owning households. However, the magnitude of the association was not large (i.e., about 20% more common). Our findings add to the body of evidence suggesting that increased suicide risk6 among members of firearm-owning households is not adequately explained by differential emotional distress or alcohol misuse preceding death by suicide.15
The qualitative interaction between household firearm ownership and the presence of a child in the household in relation to several suicide risk factors is noteworthy. Among individuals with children at home, there was a positive association between living in a firearm-owning household and serious mental illness. The presence of children in the home may prompt adults to acquire firearms with the perception of enhanced home defense. It is also possible that adults are cognizant of the risk that firearms pose to children in the home, potentially resulting in increased mental distress. Some respondents were likely from firearm-owning homes despite not being a firearm owner themselves; these individuals may be especially likely to have increased concern for a child’s safety.
Another possibility is that presence of children in the home may, at least partially, reflect the differential association between firearm ownership and suicide risk factors by age because respondents with children in the home tended to be younger than were those without. Although we controlled for age in adjusted analyses that included the interaction terms, our multivariable models may not have fully accounted for the relationship between age and suicide risk factors. Lastly, observed statistical interactions may be the result of chance findings and not genuine modifications of effects. Regardless, the presence of children in the home as a modifier of the relationship between mental health and firearm ownership warrants further study.
In examining storage practices among firearm-owning households, we found no significant differences in mental health indicators by storage practices. However, alcohol misuse was more commonly reported by individuals from households with at least 1 locked and loaded or unlocked and unloaded firearm and most prevalent among adults living in a home with unlocked and loaded firearms. A meta-analysis found that individuals who consumed alcohol were more likely to carry out a suicide with a firearm than were those who did not.30 That this suicide risk factor is most prevalent among individuals in homes with the least safe storage practices is a cause for concern. Because of the impulsivity that contributes to death by suicide, even the extra minutes required to unlock and load a safely stored firearm may prevent some deaths by suicide.
Limitations
There are limitations to this analysis. As with other survey measures, data may be biased to the extent by which the characteristics of individuals whose records we included differed from those of individuals whose records we excluded. It has also been previously documented that reported household firearm ownership and storage practices are influenced by characteristics of survey respondents.31 Gender is associated with reporting household firearm ownership. In data from the General Social Survey from 1980 to 1996, estimates of firearm prevalence varied by a median of 7 percentage points in married households, with husbands reporting household firearm ownership more frequently than did wives.32 This pattern is present in our data; among married survey respondents, men were more likely than were women to report the presence of a firearm.
Reported storage practices also vary by respondent characteristics. Research using Oregon State BRFSS data from 1992 to 1993 found that individuals who reported having never used household firearms were less likely to indicate that firearms were sometimes or always stored loaded (17.5% of nonusers vs 32.2% of users) and that firearms were stored loaded and unlocked some or all of the time (12.4% of nonusers vs 23.6% of users).33 Although the current Washington BRFSS questionnaire does not inquire about firearm use, who within the household is the firearm owner, or reasons for firearm ownership, unsafe storage practices are likely underreported because of variation in use of household firearms and social desirability bias.
Ascertaining mental health status requires asking sensitive questions. Previous research has found that a greater prevalence of poor mental health is reported in audio computer-assisted self-interviewing than in interviewer-administered paper-and-pencil interviews.34 Although BRFSS does not involve face-to-face interviewing, interaction with an interviewer is less private than are self-administered questionnaires. This introduces the opportunity for bias via underreporting of depression and other mental health indicators. Previous evaluation of BRFSS data has also suggested that individuals with severe depression are less likely to provide complete information on their mental health status.35 Additionally, for our indicators of interest, respondents to BRFSS reported only their own health status, whereas the measures of firearm ownership and storage were for the household. Individuals may report good mental health status, but this may not represent the mental health of all household members. BRFSS did not ascertain other indicators of mental health, such as relationship fulfillment and social support.
Public Health Implications
Many states have not inquired about firearm ownership and storage practices in BRFSS for more than a decade. Therefore, previous published investigations on these measures have relied on older data or data with smaller sample sizes. The BRFSS sampling frame provided a large, representative sample of Washington State adults for this analysis.
Our findings provide the most recent evidence suggesting that increased suicide risks among firearm-owning household members are not attributable to a greater underlying risk for suicidality. This evidence has public health and clinical implications. A nationally representative survey of adults in the United States found that 2 in 3 adults agree that it is sometimes appropriate for clinicians to discuss firearms with their patients.36 Providers have been recommended to discuss household firearm ownership and storage practices with their patients. Our findings on greater prevalence of alcohol misuse among individuals in households with unsafe storage practices than among their counterparts from households with safe storage practices highlight the importance of such recommendations.
Our findings also support initiating lethal means safety interventions on a larger scale. Recent research found that only around 15% of firearm owners, and an even smaller percentage of nonowners living in firearm-owning households, have had formal training in suicide prevention,37 indicating the importance of enhancing awareness about firearm-related suicide. There are also growing community-based efforts to promote safe firearm storage. The preliminary evaluation of 1 such intervention in King County, Washington, was successful in improving storage habits among attendees.38 Furthermore, the US Government Accountability Office recently released a report on the effectiveness of safe storage campaigns.39 Firearm storage practices may be improved on a larger scale with the expansion of such programs. Our results further support the importance of developing and testing such interventions.
ACKNOWLEDGMENTS
Firearms-related state-added questions were funded by Clark County Public Health, Kitsap Public Health District, Public Health-Seattle & King County, Snohomish Health District, Spokane Regional Health District, Tacoma-Pierce County Health Department, and the Washington State Department of Health. This study was funded by Grandmothers Against Gun Violence. Behavioral Risk Factor Surveillance System data are made available by the Washington State Department of Health, Center for Health Statistics, Behavioral Risk Factor Surveillance System and supported in part by the Centers for Disease Control and Prevention (cooperative agreements U58/SO000047-3 [2013]; U58/DP006066-01 [2015]; and NU58/DP006066-02-02 [2016]).
The authors are thankful for the contribution of colleagues at Public Health-Seattle & King County and the Washington Department of Health. The authors also thank the directors and leadership of the local and state public health department for their dedicated support and encouragement.
HUMAN PARTICIPANT PROTECTION
The Washington State and University of Washington institutional review boards have determined that the Behavioral Risk Factor Surveillance System is exempt from review and approval.
Footnotes
REFERENCES
- 1.Drapeau CW, McIntosh JL. U.S.A. suicide 2015: official final data. 2016. Available at: http://www.suicidology.org/portals/14/docs/resources/factsheets/2015/2015datapgsv1.pdf?ver=2017-01-02-220151-870. Accessed November 26, 2017.
- 2.Miller M, Lippmann SJ, Azrael D, Hemenway D. Household firearm ownership and rates of suicide across the 50 United States. J Trauma. 2007;62(4):1029–1034. doi: 10.1097/01.ta.0000198214.24056.40. discussion 1034–1035. [DOI] [PubMed] [Google Scholar]
- 3.Hemenway D, Miller M. Association of rates of household handgun ownership, lifetime major depression, and serious suicidal thoughts with rates of suicide across US census regions. Inj Prev. 2002;8(4):313–316. doi: 10.1136/ip.8.4.313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Miller M, Azrael D, Barber C. Suicide mortality in the United States: the importance of attending to method in understanding population-level disparities in the burden of suicide. Annu Rev Public Health. 2012;33:393–408. doi: 10.1146/annurev-publhealth-031811-124636. [DOI] [PubMed] [Google Scholar]
- 5.Kellermann AL, Rivara FP, Somes G et al. Suicide in the home in relation to gun ownership. N Engl J Med. 1992;327(7):467–472. doi: 10.1056/NEJM199208133270705. [DOI] [PubMed] [Google Scholar]
- 6.Anglemyer A, Horvath T, Rutherford G. The accessibility of firearms and risk for suicide and homicide victimization among household members: a systematic review and meta-analysis. Ann Intern Med. 2014;160(2):101–110. doi: 10.7326/M13-1301. [DOI] [PubMed] [Google Scholar]
- 7.National Research Council. Firearms and Violence: A Critical Review. Washington, DC: National Academies Press; 2005. [Google Scholar]
- 8.Ilgen MA, Zivin K, McCammon RJ, Valenstein M. Mental illness, previous suicidality, and access to guns in the United States. Psychiatr Serv. 2008;59(2):198–200. doi: 10.1176/ps.2008.59.2.198. [DOI] [PubMed] [Google Scholar]
- 9.Miller M, Barber C, Azrael D, Hemenway D, Molnar BE. Recent psychopathology, suicidal thoughts and suicide attempts in households with and without firearms: findings from the National Comorbidity Study Replication. Inj Prev. 2009;15(3):183–187. doi: 10.1136/ip.2008.021352. [DOI] [PubMed] [Google Scholar]
- 10.Simonetti JA, Mackelprang JL, Rowhani-Rahbar A, Zatzick D, Rivara FP. Psychiatric comorbidity, suicidality, and in-home firearm access among a nationally representative sample of adolescents. JAMA Psychiatry. 2015;72(2):152–159. doi: 10.1001/jamapsychiatry.2014.1760. [DOI] [PubMed] [Google Scholar]
- 11.Oslin DW, Zubritsky C, Brown G, Mullahy M, Puliafico A, Ten Have T. Managing suicide risk in late life access to firearms as a public health risk. Am J Geriatr Psychiatry. 2004;12(1):30–36. [PubMed] [Google Scholar]
- 12.Morrissey TW. Parents’ depressive symptoms and gun, fire, and motor vehicle safety practices. Matern Child Health J. 2016;20(4):799–807. doi: 10.1007/s10995-015-1910-z. [DOI] [PubMed] [Google Scholar]
- 13.Wintemute GJ. Association between firearm ownership, firearm-related risk and risk reduction behaviours and alcohol-related risk behaviours. Inj Prev. 2011;17(6):422–427. doi: 10.1136/ip.2010.031443. [DOI] [PubMed] [Google Scholar]
- 14.Sorenson SB, Vittes KA. Mental health and firearms in community-based surveys. Eval Rev. 2008;32(3):239–256. doi: 10.1177/0193841X08315871. [DOI] [PubMed] [Google Scholar]
- 15.Miller M, Swanson SA, Azrael D. Are we missing something pertinent? A bias analysis of unmeasured confounding in the firearm-suicide literature. Epidemiol Rev. 2016;38(1):62–69. doi: 10.1093/epirev/mxv011. [DOI] [PubMed] [Google Scholar]
- 16.Shenassa ED, Rogers ML, Spalding KL, Roberts MB. Safer storage of firearms at home and risk of suicide: a study of protective factors in a nationally representative sample. J Epidemiol Community Health. 2004;58(10):841–848. doi: 10.1136/jech.2003.017343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Grossman DC, Mueller BA, Riedy C et al. Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA. 2005;293(6):707–714. doi: 10.1001/jama.293.6.707. [DOI] [PubMed] [Google Scholar]
- 18.Simonetti JA, Theis MK, Rowhani-Rahbar A et al. Firearm storage practices in households of adolescents with and without mental illness. J Adolesc Health. 2017;61(5):583–590. doi: 10.1016/j.jadohealth.2017.05.017. [DOI] [PubMed] [Google Scholar]
- 19.Anestis MD, Anestis JC. Suicide rates and state laws regulating access and exposure to handguns. Am J Public Health. 2015;105(10):2049–2058. doi: 10.2105/AJPH.2015.302753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? J Clin Psychiatry. 2009;70(1):19–24. [PubMed] [Google Scholar]
- 21.Barber CW, Miller MJ. Reducing a suicidal person’s access to lethal means of suicide: a research agenda. Am J Prev Med. 2014;47(3 suppl 2):S264–S272. doi: 10.1016/j.amepre.2014.05.028. [DOI] [PubMed] [Google Scholar]
- 22.Miller M, Barber C, White RA, Azrael D. Firearms and suicide in the United States: is risk independent of underlying suicidal behavior? Am J Epidemiol. 2013;178(6):946–955. doi: 10.1093/aje/kwt197. [DOI] [PubMed] [Google Scholar]
- 23.Siegel M, Ross CS, King C., 3rd A new proxy measure for state-level gun ownership in studies of firearm injury prevention. Inj Prev. 2014;20(3):204–207. doi: 10.1136/injuryprev-2013-040853. [DOI] [PubMed] [Google Scholar]
- 24.Cwik MF, Tingey L, Maschino A et al. Decreases in suicide deaths and attempts linked to the White Mountain Apache Suicide Surveillance and Prevention System, 2001–2012. Am J Public Health. 2016;106(12):2183–2189. doi: 10.2105/AJPH.2016.303453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Stein EM, Gennuso KP, Ugboaja DC, Remington PL. The epidemic of despair among White Americans: trends in the leading causes of premature death, 1999–2015. Am J Public Health. 2017;107(10):1541–1547. doi: 10.2105/AJPH.2017.303941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Siegel M, Rothman EF. Firearm ownership and suicide rates among US men and women, 1981–2013. Am J Public Health. 2016;106(7):1316–1322. doi: 10.2105/AJPH.2016.303182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kessler RC, Barker PR, Colpe LJ et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry. 2003;60(2):184–189. doi: 10.1001/archpsyc.60.2.184. [DOI] [PubMed] [Google Scholar]
- 28.Moriarty DG, Zack MM, Kobau R. The Centers for Disease Control and Prevention’s Healthy Days Measures—population tracking of perceived physical and mental health over time. Health Qual Life Outcomes. 2003;1:37. doi: 10.1186/1477-7525-1-37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Office of Disease Prevention and Health Promotion. Dietary guidelines for Americans: 2015–2020, eighth edition. 2015. Available at: http://health.gov/dietaryguidelines/2015/guidelines. Accessed September 16, 2017.
- 30.Branas CC, Han S, Wiebe DJ. Alcohol use and firearm violence. Epidemiol Rev. 2016;38(1):32–45. doi: 10.1093/epirev/mxv010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Azrael D, Miller M, Hemenway D. Are household firearms stored safely? It depends on whom you ask. Pediatrics. 2000;106(3):E31. doi: 10.1542/peds.106.3.e31. [DOI] [PubMed] [Google Scholar]
- 32.Ludwig J, Cook PJ, Smith TW. The gender gap in reporting household gun ownership. Am J Public Health. 1998;88(11):1715–1718. doi: 10.2105/ajph.88.11.1715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Nelson DE, Powell K, Johnson CJ, Mercy J, Grant-Worley JA. Household firearm storage practices: do responses differ by whether or not individuals ever use firearms? Am J Prev Med. 1999;16(4):298–302. doi: 10.1016/s0749-3797(98)00096-8. [DOI] [PubMed] [Google Scholar]
- 34.Epstein JF, Barker PR, Kroutil LA. Mode effects in self-reported mental health data. Public Opin Q. 2001;65(4):529–549. [Google Scholar]
- 35.Frankel MR, Battaglia MP, Balluz L, Strine T. When data are not missing at random: implications for measuring health conditions in the Behavioral Risk Factor Surveillance System. BMJ Open. 2012;2(4):e000696. doi: 10.1136/bmjopen-2011-000696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Betz ME, Azrael D, Barber C, Miller M. Public opinion regarding whether speaking with patients about firearms is appropriate. Ann Intern Med. 2016;165(8):543–550. doi: 10.7326/M16-0739. [DOI] [PubMed] [Google Scholar]
- 37.Rowhani-Rahbar A, Lyons VH, Simonetti JA, Azrael D, Miller M. Formal firearm training among adults in the USA: results of a national survey. Inj Prev. 2018;24(2):161–165. doi: 10.1136/injuryprev-2017-042352. [DOI] [PubMed] [Google Scholar]
- 38.Simonetti JA, Rowhani-Rahbar A, King C, Bennett E, Rivara FP. Evaluation of a community-based safe firearm and ammunition storage intervention. Inj Prev. 2017 doi: 10.1136/injuryprev-2016-042292. injuryprev-2016-042292. [DOI] [PubMed] [Google Scholar]
- 39.Government Accountability Office. Personal firearms: programs that promote safe storage and research on their effectiveness. 2017. Available at: https://www.gao.gov/products/GAO-17-665. Accessed April 30, 2018.