Lethality of Civilian Active Shooter Incidents With and Without Semiautomatic Rifles in the United States

Semiautomatic rifles have been used in some of the largest active shooter incidents in US history.¹ Many weapons were banned in 1994 under the federal assault weapons ban but were reintroduced to the public marketplace in 2004.² Currently, there are no comprehensive assessments of injuries from different types of firearms. We compared the number of persons wounded, killed, and either wounded or killed during active shooter incidents with and without semiautomatic rifles.

An active shooter incident is defined by the Federal Bureau of Investigation (FBI) as a situation in which an individual is actively engaged in killing or attempting to kill people in a confined or populated area.³ The FBI has tracked all active shooter incidents since 2000 and has the most comprehensive data set available.³ We retrieved active shooter incident characteristics from the publicly accessible FBI database through 2017 (accessed May 18, 2018).³ For each incident, we extracted shooter age, name, year, location (city and state), number of people wounded, killed, and wounded or killed, place of shooting (commerce, education, government, open space, residences, health care, and house of worship), and type of firearms present (rifle, shotgun, handgun).

The FBI reports do not distinguish whether a rifle was semiautomatic; therefore, for each incident in which the FBI reported that a rifle was present, a media content analysis was performed to identify semiautomatic rifle presence. An a priori search hierarchy was established in which the primary data sources were court and police documents or statements (44.9%; 35 of 78), and secondary data sources were news articles. At least 3 news articles from different media outlets were required to triangulate data. No discrepancies among sources were found. All incidents with the presence of a semiautomatic rifle were classified as semiautomatic rifle incidents regardless of other firearm presence. The Las Vegas, Nevada, shooting, which represented a statistical outlier, and the San Bernardino, California, shooting, which had more than 1 shooter present, were excluded. Negative binomial regression was used to estimate the association between presence of a semiautomatic rifle and the total numbers nonfatally wounded, killed, and either wounded or killed, and the percentage of persons who died if wounded at the incident, controlling for the place and year of shooting and the presence of other firearms. Significance was set at P < .05 (2-sided). Stata version 15.1 was used for analysis.

Of the 248 active shooter incidents, 76 involved a rifle, and we identified the type in all instances. A semiautomatic rifle was involved in 24.6% (n = 61) of incidents, and 75.4% (n = 187) involved handguns (n = 154), shotguns (n = 38), and non–semiautomatic rifles (n = 15). Multiple firearm types were involved in 60.7% (n = 37 of 61) of semiautomatic rifle incidents and 25.1% (n = 47) of non–semiautomatic rifle incidents.

There were 898 persons wounded and 718 killed. Active shooter incidents with vs without the presence of a semiautomatic rifle were associated with a higher incidence of persons wounded (unadjusted mean, 5.48 vs 3.02; incidence rate ratio [IRR], 1.81 [95% CI, 1.30-2.53]), killed (mean, 4.25 vs 2.49; IRR, 1.97 [95% CI, 1.38-2.80]), and wounded or killed (mean, 9.72 vs 5.47; IRR, 1.91 [95% CI, 1.46-2.50]) (Figure). The percentage of persons who died if wounded in incidents with a semiautomatic rifle (43.7% [n = 259 of 593]) was similar to the percentage who died in incidents without a semiautomatic rifle (44.9% [n = 459 of 1023]) (IRR, 0.99 [95% CI, 0.60-1.61]).

Although 44% of persons wounded in active shooter incidents died of their injuries, irrespective of the type of firearm used, more people were wounded and killed in incidents in which semiautomatic rifles were used compared with incidents involving other firearms. Semiautomatic rifles are designed for easy use, can accept large magazines, and fire high-velocity bullets, enabling active shooters to wound and kill more people per incident.⁴

Limitations of this study include the lack of data on specific injuries, demographics, and other details of the incidents. Incidents involving semiautomatic rifles may differ from other incidents in ways that may partially explain the association but could not be controlled (ie, intentionality of the shooter). This lack of data highlights the need for a national centralized database to inform the debate on an assault weapons ban.

Accepted for Publication: July 11, 2018.

Correction: This article was corrected on November 27, 2018, to improve an unclear sentence and revise an incorrect number in the figure.

Corresponding Author: Adil H. Haider, MD, MPH, Center for Surgery and Public Health, Brigham and Women’s Hospital, 75 Francis St, Boston, MA 02115 (ahhaider@bwh.harvard.edu).

Author Contributions: Dr Haider 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: De Jager, Goralnick, McCarty, Hashmi, Haider.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: De Jager, Goralnick, McCarty, Haider.

Critical revision of the manuscript for important intellectual content: De Jager, Goralnick, McCarty, Hashmi, Jarman, Haider.

Statistical analysis: Goralnick, McCarty, Hashmi, Jarman, Haider.

Administrative, technical, or material support: De Jager, Goralnick, McCarty, Haider.

Supervision: Goralnick, Haider.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Haider reports stock holdings and being cofounder of Patient Doctor Technologies. No other disclosures were reported.