Public Release: 

The Lancet: Universal background checks for purchasing guns and ammunition could substantially cut gun deaths in the USA

The Lancet

A nationwide study analysing gun-control laws in the USA has found that just nine of the 25 state laws are effective in reducing firearm deaths. The research, published in The Lancet, suggests that if all US states were to expand the three laws that have the strongest effect on gun deaths--universal background checks for purchasing guns and ammunition, and firearm identification [1]--the national rate of gun deaths could be cut by over 90%.

"Our study is the first to examine the impact of specific gun laws on gun-related deaths across the USA while taking account of a range of other factors such as gun ownership and unemployment" says lead author Dr Bindu Kalesan from Boston University, Boston, USA. "The findings suggest that very few of the existing state gun-control laws actually reduce gun deaths, highlighting the importance of focusing on relevant and effective gun legislation. Background checks for all people buying guns and ammunition, including private sales, are the most effective laws we have to reduce the number of gun deaths in the USA,"[2].

More than 90 people are killed every day by guns in the USA. In 2010, 31672 gun deaths were recorded, equivalent to 10.1 deaths per 100000 people. Hawaii recorded the lowest rate of gun deaths (45 deaths per year) at 3.31 per 100000 citizens, while Alaska (144) topped the table at 20.3 per 100000 (see appendix table 2).

The link between state levels of gun ownership and gun deaths is well established. But less is known about the effectiveness of existing gun laws. US states have introduced a range of gun laws to strengthen or deregulate the main federal gun control law, the Brady Law, which requires background checks for gun purchases from a federally-licensed dealer. However, around 40% of all gun sales in America are estimated to be private transactions that do not require background checks.

In this study, Dr Kalesan and colleagues examined the relationship between gun-control laws and recorded gun deaths across all US states using data from the Center for Disease Control and Prevention's Injury Statistics Query and Reporting System. They used a model fitted with data on 25 different gun-control laws in each state (table 1), and rates of gun deaths, gun ownership, gun export (a proxy of the amount of firearms in each state), non-firearm homicides (a measure of crime in each state), and unemployment to calculate which particular gun laws might impact firearm deaths across the USA in 2010.

The findings show that nine laws are associated with a reduced likelihood of gun deaths, nine with increased gun deaths, and seven did not show any conclusive association (figure page 4). For example, laws that restrict firearm access to children (eg, locks and age restrictions) were shown to be ineffective, while stand-your-ground laws that allow an individual to use deadly force in self-defence significantly increased gun-related deaths. These findings persisted even after removing the effect of other factors that might affect gun deaths such unemployment and gun exports.

The model predicts that a federal law expanding background checks for all gun purchases could more than halve the national gun death rate from 10.35 to 4.46 per 100000 people (57% reduction). Similarly, background checks for all ammunition purchases could cut overall gun deaths to 1.99 per 100000 (81% decrease), and firearm identification could reduce deaths to 1.81 per 100000 (83% reduction) (table 3 page 6) [3]. Federal implementation of all three laws could reduce national overall gun deaths to 0·16 per 100 000 (over 90% reduction). However, the authors caution that they expect the fall in deaths to be a long-term effect that could take many years to achieve.

In a linked Comment highlighting the limitations of the study, David Hemenway, Professor of Public Policy at Harvard School of Public Health, Boston, USA cautions that while the investigators tried to control for a number of variables, "many other important factors were not controlled for (eg, poverty, alcohol consumption, urbanicity, and mental health). With data for just 1 year, the authors could not examine the rates of firearm fatalities before and after the passage of any law."

He adds, "Kalesan and colleagues' findings also suggest some possible statistical problems. Their results suggest that if all US states would merely implement universal background checks for firearm and ammunition purchase, national rates of firearm mortality would decrease by more than 90%. That result is too large--if only firearm suicide and firearm homicide could be reduced so easily...Although not the final word, the study by Kalesan and colleagues is a step in the right direction of trying to bring more scientific evidence to bear on the types of policies that could be most effective in reducing the serious gun-violence problem in the USA.



[1] To help to keep guns out of the hands of individuals with a history of violent crime, domestic violence, substance abuse, and severe mental illness, and others who are prohibited from owning guns, licensed firearm dealers are unable to sell a gun until a background check is performed by the National Instant Criminal Background Check System, a division of the FBI. Firearm identification by either microstamping or ballistic fingerprinting of semi-automatic handguns enable cartridge cases recovered at crime scenes to be linked to the particular gun that fired them. Ballistic fingerprinting requires that firearm manufacturers store images of cartridge cases markings allowing for gun identification. Microstamping involves laser engraving a serial number onto cartridges after a gun is fired.

[2] Quote direct from author and cannot be found in text of Article.

[3] Data for the predicted risk of death associated with protective firearm laws in all 50 US states in 2010 are in the appendix table 2.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.