Abstract
This analysis of the Global Terrorism Database evaluates differences in the use of firearms in terrorist attacks between the United States, Canada, Europe, Australia, and New Zealand.
Although firearms are used in only a small proportion of terrorist attacks, these highly publicized events shed light on access to weapons and mass shootings. The sociopolitical and cultural context surrounding firearms, including the proportion of individuals owning guns, varies between countries. The United States has a higher rate of firearms deaths than other high-income countries. We compared the proportion of terrorist attacks committed with firearms in the United States with the proportion in other high-income countries. We also compared the lethality of attacks with firearms to those by other means.
Methods
We queried The Global Terrorism Database from 2002 to 2016. Maintained by the National Consortium for the Study of Terrorism and Responses to Terrorism at the University of Maryland, this database incorporates a methodology that includes both machine learning and manual review to abstract high-quality information from more than 1 million daily media reports published worldwide in over 80 languages. The database defines a terrorist attack as the “use of illegal force and violence by a non-state actor to attain a political, economic, religious, or social goal through fear, coercion, or intimidation.” For each attack, the location, type, and number of fatalities are collected. The database categorizes weapons as biological, chemical, explosive, fake weapons, firearms, incendiary (eg, arson), melee, sabotage equipment, vehicle (nonexplosive), other, and unknown.
To avoid calculating proportions in countries with few data points, we calculated the proportion of attacks involving firearms among countries in the top 75th percentile (10 attacks or more) over the study period. The number of fatalities per attack was calculated by weapon type. Of the 23 countries with at least 1 attack, 17 were in the upper 75th percentile of total attacks; the countries are listed in footnote “d” in the Table. We estimated incidence rate ratios (IRRs) and 95% confidence intervals (CIs) per fatality in the attack with a Poisson distribution for attacks using firearms, vehicle/melee, incendiary, or biological/chemical weapons compared with explosives.
Table. Count of Total Attacks and Type of Weapon Used by Country, Group, or Region and Yeara.
| Year | Total Attacks (N = 2817) |
Firearms (n = 260, 9.2%) |
Explosives (n = 1379, 49.0%) |
Incendiary (n = 1015, 36.0%) |
Vehicle/Melee (n = 151, 5.4%) |
Miscellaneousb (n = 88, 3.1%) |
|---|---|---|---|---|---|---|
| United States | ||||||
| 2002 | 33 | 2 | 20 | 10 | 2 | 0 |
| 2003 | 32 | 0 | 2 | 26 | 2 | 1 |
| 2004 | 9 | 0 | 0 | 8 | 0 | 1 |
| 2005 | 20 | 1 | 5 | 14 | 0 | 0 |
| 2006 | 6 | 1 | 1 | 3 | 2 | 0 |
| 2007 | 9 | 0 | 2 | 6 | 0 | 1 |
| 2008 | 18 | 1 | 3 | 12 | 2 | 1 |
| 2009 | 10 | 5 | 2 | 1 | 2 | 1 |
| 2010 | 17 | 6 | 4 | 4 | 1 | 1 |
| 2011 | 10 | 1 | 3 | 5 | 0 | 1 |
| 2012 | 19 | 3 | 1 | 15 | 0 | 0 |
| 2013 | 20 | 4 | 7 | 6 | 0 | 2 |
| 2014 | 26 | 18 | 4 | 4 | 1 | 0 |
| 2015 | 39 | 11 | 8 | 18 | 4 | 1 |
| 2016 | 61 | 14 | 7 | 32 | 7 | 1 |
| Total | 329 | 67 | 69 | 164 | 23 | 11 |
| Canada | ||||||
| 2002 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2003 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2004 | 2 | 0 | 1 | 1 | 0 | 0 |
| 2005 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2006 | 2 | 0 | 0 | 2 | 0 | 0 |
| 2007 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2008 | 5 | 0 | 3 | 0 | 2 | 0 |
| 2009 | 4 | 0 | 3 | 1 | 0 | 0 |
| 2010 | 2 | 0 | 2 | 0 | 0 | 0 |
| 2011 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2012 | 3 | 0 | 1 | 1 | 0 | 1 |
| 2013 | 4 | 0 | 2 | 2 | 0 | 0 |
| 2014 | 3 | 2 | 0 | 0 | 1 | 0 |
| 2015 | 5 | 0 | 2 | 2 | 1 | 0 |
| 2016 | 6 | 1 | 1 | 2 | 1 | 0 |
| Total | 36 | 3 | 15 | 11 | 5 | 1 |
| Western Europec | ||||||
| 2002 | 120 | 7 | 87 | 27 | 3 | 0 |
| 2003 | 121 | 5 | 83 | 21 | 3 | 0 |
| 2004 | 59 | 5 | 47 | 4 | 4 | 0 |
| 2005 | 100 | 4 | 74 | 22 | 1 | 0 |
| 2006 | 98 | 7 | 63 | 28 | 1 | 0 |
| 2007 | 72 | 10 | 48 | 12 | 2 | 0 |
| 2008 | 163 | 5 | 72 | 78 | 10 | 0 |
| 2009 | 181 | 19 | 54 | 107 | 5 | 0 |
| 2010 | 133 | 13 | 93 | 23 | 3 | 0 |
| 2011 | 93 | 10 | 60 | 18 | 1 | 0 |
| 2012 | 190 | 9 | 155 | 24 | 4 | 0 |
| 2013 | 254 | 15 | 160 | 70 | 11 | 13 |
| 2014 | 215 | 8 | 108 | 93 | 7 | 0 |
| 2015 | 335 | 38 | 111 | 161 | 25 | 18 |
| 2016 | 269 | 31 | 76 | 121 | 41 | 17 |
| Total | 2403 | 186 | 1291 | 809 | 121 | 48 |
| Australia/New Zealand | ||||||
| 2002 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2003 | 3 | 0 | 0 | 0 | 0 | 0 |
| 2004 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2005 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2006 | 2 | 0 | 0 | 1 | 0 | 1 |
| 2007 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2008 | 8 | 0 | 2 | 1 | 0 | 0 |
| 2009 | 1 | 0 | 1 | 0 | 0 | 0 |
| 2010 | 1 | 1 | 0 | 0 | 0 | 0 |
| 2011 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2012 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2013 | 1 | 0 | 0 | 1 | 0 | 0 |
| 2014 | 9 | 2 | 0 | 6 | 1 | 0 |
| 2015 | 14 | 1 | 1 | 12 | 0 | 0 |
| 2016 | 10 | 0 | 0 | 10 | 1 | 0 |
| Total | 49 | 4 | 4 | 31 | 2 | 1 |
All data are from the Global Terrorism Database. In addition, totals may not equal sum of counts by weapon type because single attacks can have multiple types of weapons used, and totals include attacks using biological/chemical weapons (n = 37, 1.3%), which are not reported in the table.
Includes fake weapons, sabotage, and weapon categories “Other” and “Unknown.”
Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Malta, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, and Vatican City.
Results
From 2002 through 2016, the database captured 2817 terrorist attacks in the United States, Canada, Western Europe, Australia, and New Zealand, of which 2403 (85.3%) were in Western Europe and 329 (11.7%) in the United States. The Table details the count of attacks by weapon used in country or region and year. Nearly all attacks (n = 2788, 99.0%) occurred in countries with 10 or more attacks. The types of attacks were as follows: explosives (n = 1379, 49.0%), incendiary (n = 1015, 36.0%), firearms (n = 260, 9.2%), vehicle/melee (n = 151, 5.4%), and miscellaneous (n = 88, 3.1%). In 2015 and 2016, compared with earlier years, there were notable increases in attacks involving vehicle/melee in the United States and Western Europe.
The Figure, A shows the proportion of firearms attacks among countries with 10 or more attacks, with the United States at 20.4% (n = 67) followed by the Netherlands at 14.3% (n = 3) (χ2 for all countries, P < .001). In the 2817 attacks, there were 1031 fatalities, of which 566 (54.9%) were attributed to firearms. The Figure, B shows the number of fatalities per attack by weapon type. Among all weapon types compared with explosives, the IRR per fatality was 4.75 (95% CI, 4.18-5.39) for attacks with firearms, 1.21 (95% CI, 0.91-1.59) for vehicle/melee, and 0.05 (95% CI, 0.03-0.08) for incendiaries.
Figure. Proportion of Firearms Attacks and Fatalities per Attack, 2002-2016.
A, Proportion of attacks in the upper 75th percentile of total attacks. B, Fatalities from all attacks. All data are from the Global Terrorism Database.
Discussion
Although firearms were used in fewer than 10% of terrorist attacks between 2002 and 2016, they accounted for about 55% of the fatalities. Among countries with 10 or more attacks, the proportion involving firearms in the United States was higher than in any other nation. Moreover, the risk of fatality was considerably higher in attacks committed with firearms than in attacks using other weapons. In the United States and other countries, government policies and legislative efforts to protect citizens from terrorism should consider the proportions and lethality of terrorist attacks committed with firearms.
References
- 1.Rosenbaum JE. Gun utopias? firearm access and ownership in Israel and Switzerland. J Public Health Policy. 2012;33(1):46-58. doi: 10.1057/jphp.2011.56 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Richardson EG, Hemenway D. Homicide, suicide, and unintentional firearm fatality: comparing the United States with other high-income countries, 2003. J Trauma. 2011;70(1):238-243. doi: 10.1097/TA.0b013e3181dbaddf [DOI] [PubMed] [Google Scholar]
- 3.Grinshteyn E, Hemenway D. Violent death rates: the US compared with other high-income OECD countries, 2010. Am J Med. 2016;129(3):266-273. doi: 10.1016/j.amjmed.2015.10.025 [DOI] [PubMed] [Google Scholar]
- 4.University of Maryland Global Terrorism Database. https://www.start.umd.edu/gtd/. Accessed May 15, 2017.