Sociodemographic Factors Associated With High Risk for Firearm Suicide Among US Adults

Although firearms are a highly lethal means of attempting suicide¹ and were used in 24 432 of 48 344 US suicide deaths (50.5%) in 2018, little is known about factors associated with firearm suicide risk. Much of the previous research describing firearm suicide compared suicide decedents using firearms with other methods.² Without prospectively evaluating firearm suicide risk among living populations, however, these analyses did not yield firearm suicide rates and thus have limited utility for identifying high-risk groups. To increase our understanding of firearm suicide risk patterns, we compared firearm suicide risk across sociodemographic groups of US adults. Results might help inform tailoring firearm safety interventions to high-risk groups, such as the lethal means safety counseling widely available within Veterans Health Administration facilities.

This cohort study used data from respondents in the Mortality Disparities Across Communities³ study, which linked 2008 American Community Survey data to 2008-2015 National Death Index death records. Respondents were classified by age at survey, sex, race/ethnicity, marital status, educational level, employment, income, disability, military service, residence, and US citizenship. State-level data on household firearms were obtained from the 2004 Behavioral Risk Factor Surveillance System.⁴ International Statistical Classification of Diseases, Tenth Revision, Clinical Modification diagnosis codes identified firearm suicide (X72-X74) and nonfirearm suicide (X60-X71, X75-X84, Y87.0, U03) deaths. For each characteristic, we assessed the number of respondents, follow-up years, and mortality rates per 100 000 person-years. Hazard ratios (HRs) adjusted for age, sex, and race/ethnicity. Separate analyses were also performed for men and women. The US Census Bureau’s Disclosure Review Board approved the study procedures including a waiver of informed consent under 45 CFR 46.116 and approved the data output.

Firearm suicide rates were highest among individuals with military service (21.2 per 100 000 person-years; 95% CI, 19.4-23.1 per 100 000 person-years), disabled individuals (14.0 per 100 000 person-years; 95% CI, 12.7-15.3 per 100 000 person-years), and men (13.6 per 100 000 person-years; 95% CI, 13.0-14.3 per 100 000 person-years) (Table 1). Other high-risk groups included adults aged 65 years or older, rural residents, adults living in states with more than 50% firearm ownership, and separated or divorced persons. Compared with noncitizens, US citizens had a greater risk for firearm suicide (HR, 3.00; 95% CI, 2.50-4.39) and nonfirearm suicide (HR, 1.20; 95% CI, 0.97-1.48).

Among men, firearm suicide rates were highest for widowers (29.1 per 100 000 person-years; 95% CI, 22.9-36.5 per 100 000 person-years), adults aged 65 years or older (25.9 per 100 000 person-years; 95% CI, 23.4-28.5 per 100 000 person-years), and rural residents (25.7 per 100 000 person-years; 95% CI, 23.2-28.3 per 100 000 person-years) (Table 2). Male firearm suicide rates were 19.9 per 100 000 person-years) (95% CI, 16.2-24.1 per 100 000 person-years) in states with high firearm ownership and 7.9 per 100 000 person-years (95% CI, 7.1-8.9 per 100 000 person-years) in states with low firearm ownership. Rates of nonfirearm suicide were 6.0 per 100 000 person-years (95% CI, 4.1-8.6 per 100 000 person-years) in states with high firearm ownership and 9.8 per 100 000 person-years (95% CI, 8.9-10.9 per 100 000 person-years) in states with low firearm ownership. Firearm suicide rates among women were highest for those with a net income loss (5.5 per 100 000 person-years; 95% CI, 3.2-9.0 per 100 000 person-years), those with a disability (3.8 per 100 000 person-years; 95% CI, 3.0-4.8 per 100 000 person-years), and in states with greater than 50% firearm ownership (3.5 per 100 000 person-years; 95% CI, 2.1-5.5 per 100 000 person-years).

Increased firearm suicide risk was observed among groups with high firearm ownership⁵ and groups with high overall suicide risk. The former included residents of states with high firearm ownership, rural residents, individuals with military service, men, non-Hispanic white adults, and US citizens. Adults who were older, unemployed, separated or divorced, and had a lower educational level or lower incomes have not previously been found to have elevated rates of firearm ownership but had increased firearm suicide risk in this study.

Despite evidence that firearm access is associated with increased suicide risk,⁶ firearm safety interventions are commonly criticized as leading to offsetting increases in suicide by other methods. However, greater differences in firearm suicide rates between states with high and low firearm ownership alongside more modest differences in nonfirearm suicide rates support a population-level counterfactual case against means substitution.

Limitations of the study include absence of data for several factors associated with suicide risk, such as psychiatric and medical disorders; uncertain validity of the self-reported disability measure; collection of crude state firearm ownership data in 2004 compared with 2008 to 2015; and a lack of data after 2015.

Wide distribution of firearm suicide risk across social, economic, and geographic characteristics underscores the need to develop a range of firearm safety messages for these different patient groups.

Accepted for Publication: March 16, 2020.

Corresponding Author: Mark Olfson, MD, MPH, Department of Psychiatry, Vagelos College of Physicians and Surgeons, New York State Psychiatric Institute, Columbia University, 1051 Riverside Dr, New York, NY 10032 (mo49@cumc.columbia.edu).

Published Online: May 18, 2020. doi:10.1001/jamainternmed.2020.1334

Author Contributions: Ms Cosgrove had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Olfson, Wall, Blanco.

Acquisition, analysis, or interpretation of data: Olfson, Cosgrove.

Drafting of the manuscript: Olfson, Wall.

Critical revision of the manuscript for important intellectual content: Olfson, Cosgrove, Blanco.

Statistical analysis: Cosgrove, Wall.

Administrative, technical, or material support: Cosgrove.

Conflict of Interest Disclosures: None reported.

Funding/Support: The Mortality Disparities Across Communities is supported by interagency agreements between the National Heart, Lung, and Blood Institute and the Census Bureau and between the National Institute on Aging and the Census Bureau.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Institute on Drug Abuse; the National Institutes of Health; or the United States Census Bureau.