Duke Study: Gun Laws Showed No Measurable Impact on Child Homicide
A peer-reviewed analysis of 36 state firearm laws found associations with lower youth suicide rates—but no notable reduction in firearm homicide.
Science is not supposed to protect political narratives.
It is supposed to test them.
A peer-reviewed study led by researchers from Duke University School of Medicine examined whether 36 state firearm laws were associated with lower firearm suicide and homicide rates among Americans under 18.
The researchers reviewed firearm deaths from 2009 through 2020 using mortality data from the Centers for Disease Control and Prevention and firearm-law data compiled by RAND.
The dataset included:
- 17,013 firearm deaths among minors
- 10,278 firearm homicides
- 6,735 firearm suicides
- 36 different categories of state firearm laws
The researchers found that certain laws—particularly negligent-storage child-access-prevention laws and mandatory waiting periods—were associated with lower firearm suicide rates among children.
But when they examined firearm homicide?
The researchers reported no notable differences in homicide mortality between states with and without the firearm laws studied. [1][2]
That finding deserves attention.
Not because it proves every firearm law is useless.
It does not.
It deserves attention because it directly challenges the politically convenient assumption that passing more gun laws automatically produces fewer murders.
The data did not show that.
What the Duke Researchers Actually Studied
The study, published in the Journal of the American College of Surgeons, compared childhood firearm mortality rates with the presence or absence of 36 state firearm laws.
Those laws included regulations involving:
- Background checks
- Waiting periods
- Minimum purchase and possession ages
- Child-access prevention
- Firearm storage
- Prohibited possessors
- Stand-your-ground protections
- Other restrictions on firearm possession and transfer
The researchers compared average suicide and homicide mortality rates in states with each law against rates in states without that law.
For suicide, several associations appeared.
States with negligent-storage child-access-prevention laws and waiting periods demonstrated lower youth firearm-suicide mortality. Minimum-age restrictions, however, were not associated with significant reductions in suicide mortality.
For homicide, the results were dramatically less impressive.
According to Duke Health’s summary, there were no notable distinctions between states with and without the examined firearm laws.
Lead author Dr. Krista Haines publicly acknowledged that the researchers were surprised that none of the laws appeared to affect child homicide rates.
That is not commentary from a gun-rights organization.
It is the published conclusion of researchers from a major medical institution.
The Headline Political Activists Will Not Share
When a study reports an association between a gun restriction and a reduction in deaths, the result usually becomes a headline before anyone has finished reading the abstract.
When a study finds no measurable homicide reduction, the language suddenly becomes softer.
The result is described as “mixed.”
The researchers call for more research.
The press release pivots toward demanding additional laws.
But the homicide finding remains:
Thirty-six categories of state firearm legislation produced no notable difference in child firearm-homicide mortality within this analysis.
That does not mean the researchers proved that every firearm law everywhere will always fail.
It means their data did not support the claim that the laws they studied were reliably associated with lower child firearm-homicide rates.
That distinction matters.
Science does not allow us to transform a statistically insignificant result into a successful policy simply because we approve of the policy’s intentions.
Good intentions are not outcomes.
Legislation is not evidence of effectiveness.
And counting laws is not the same as preventing murders.
What This Study Does Not Prove
Intellectual honesty cuts in both directions.
This study does not prove that “gun control never works.”
It does not examine every law, every population, every homicide or every possible implementation strategy.
It specifically examined firearm deaths among individuals younger than 18 during 2009–2020.
It was also a cross-sectional state-level database study. The published methodology describes comparisons of average mortality rates using Welch’s t-tests.
That design can identify associations—or the absence of detectable associations—but it cannot establish that a law directly caused or prevented a death.
The study also excluded states that enacted or repealed relevant laws during the study period. That creates a cleaner comparison between states with and without longstanding laws, but it removes many of the policy changes that could otherwise be studied as natural experiments.
State-level comparisons are also blunt instruments.
Two states may technically have the same law while differing substantially in:
- Enforcement
- Prosecution
- Local violent-crime patterns
- Population density
- Economic conditions
- Illegal firearm markets
- Gang and group violence
- Domestic violence
- Police staffing
- Community intervention programs
- Compliance with storage requirements
Writing a statute and enforcing it effectively are not the same thing.
Neither is enforcing a law against responsible citizens and disrupting the violent offenders who are actually driving homicide.
Why Suicide and Homicide Produced Different Results
The distinction between suicide and homicide is not an inconvenient technicality.
It is the central issue.
Suicide and homicide are different behaviors, committed under different circumstances, involving different risk factors and requiring different prevention strategies.
A waiting period may interrupt an immediate suicidal crisis by delaying access to a firearm.
Secure-storage practices may prevent an unsupervised child or distressed teenager from gaining immediate access to a household weapon.
Those are logically connected mechanisms, and the Duke findings support taking them seriously.
Homicide is different.
Firearm homicide frequently develops through interpersonal conflict, retaliatory violence, domestic abuse, criminal networks, chronic violent offending and concentrated community violence.
The CDC has repeatedly reported that firearm homicide is not evenly distributed across the population. Rates vary sharply by age, sex, geography, poverty and exposure to community violence. [3][4]
That means a law directed at a lawful retail purchase, minimum purchasing age or household storage practice may have little effect on a homicide committed by a repeat offender operating through an illegal market or violent social network.
A policy aimed at the wrong mechanism should not be expected to produce the desired result merely because it contains the word “gun.”
The Number of Laws Is Meaningless Without Measurable Results
Public safety is not a legislative collection contest.
A state does not become safer simply because its statute book becomes heavier.
The correct question is not:
How many gun laws has the government passed?
The correct questions are:
- Which specific law was intended to prevent which specific behavior?
- Was that law enforced?
- Did compliance occur?
- Did homicide decline after implementation?
- Were violent offenders removed from the streets?
- Were illegal firearm sources disrupted?
- Did the policy merely displace the method of violence?
- Were lawful citizens burdened without affecting criminal behavior?
- Can the measured outcome be attributed to the law rather than unrelated social changes?
Those questions are less emotionally satisfying than chanting “do something.”
They are also considerably more useful.
One Study Does Not Settle the Entire Debate
A broader 2026 RAND review reached different conclusions regarding several specific policies.
After reviewing a much larger body of research, RAND reported varying levels of evidence that certain policies—including child-access-prevention laws, waiting periods, domestic-violence firearm prohibitions and some background-check requirements—may affect particular homicide or suicide outcomes.
RAND also reported evidence associating some stand-your-ground and shall-issue carry laws with increased homicide or violent-crime outcomes. [5]
That broader review prevents us from honestly claiming that the Duke study ended the national debate.
But it also does not erase Duke’s null homicide findings.
The two can exist simultaneously.
Different research designs, populations, periods, policy definitions and statistical methods can produce different findings. That is exactly why serious researchers evaluate the total body of evidence instead of selecting one convenient paper and declaring victory.
The Duke study should therefore be treated neither as sacred proof that all gun laws fail nor as an embarrassing result to be buried.
It should be treated as evidence.
Important evidence.
Stop Confusing Firearm Regulation With Violence Prevention
A firearm is an instrument.
Homicide is a human behavior.
That does not make access irrelevant. It means access is only one part of a much larger causal system.
A government can regulate an instrument without successfully changing the people, relationships, criminal networks and environmental conditions producing violence.
Actual violence reduction requires far more than passing restrictions against people who were not committing the murders.
It requires:
- Identifying individuals and groups driving shootings
- Enforcing existing laws against violent offenders
- Disrupting retaliatory cycles
- Intervening before high-risk conflicts become shootings
- Prosecuting illegal trafficking and straw purchasing
- Addressing domestic violence
- Improving investigative clearance rates
- Providing credible exit routes for individuals embedded in violent networks
- Combining enforcement with community services and supervision
Focused-deterrence programs target identifiable violent groups and repeat offenders while combining law enforcement, direct communication and social-service intervention. Systematic reviews have found these strategies associated with statistically significant crime reductions, although effectiveness varies by program quality and research design. [6]
That is targeted violence prevention.
Passing another broadly written restriction and hoping criminals become impressed by the legislative effort is not a strategy.
It is political theater with a research grant attached.
Responsible Gun Owners Should Not Ignore the Suicide Findings
There is another uncomfortable truth.
Responsible gun owners should not dismiss the study simply because its homicide findings challenge gun-control narratives.
The suicide findings also matter.
Safe firearm storage is not surrender.
Preventing unauthorized access by children is not anti-Second Amendment.
Recognizing an acute suicide risk inside a household is not political weakness.
It is responsible ownership.
The Duke study found associations between certain access-control policies and lower youth suicide mortality. RAND’s broader research review likewise concluded that child-access-prevention laws reduce self-inflicted firearm injuries among young people. [1][5]
We should be capable of discussing those findings without converting responsible storage into a partisan loyalty test.
A serious gun owner controls access to the firearm.
A serious policymaker demands evidence before restricting constitutional conduct.
Those positions are not contradictory.
The Real Scientific Conclusion
The Duke study does not scientifically justify the sweeping headline that every gun-control law has no effect on every homicide.
But it supports a conclusion that should make policymakers uncomfortable:
The existence of more firearm laws does not guarantee lower firearm-homicide rates.
In this study, 36 categories of state firearm laws were compared with more than 10,000 child firearm homicides.
The researchers found no notable homicide differences between states with and without those laws.
That should force a change in the conversation.
Stop measuring commitment by the number of laws passed.
Stop treating lawful firearm ownership as a substitute variable for violent criminal behavior.
Stop pretending homicide and suicide are interchangeable because both may involve a firearm.
Stop calling a policy successful before outcomes prove it.
And stop using the word “science” as decoration for a conclusion selected in advance.
The objective is not to pass gun laws.
The objective is to prevent deaths.
Those are not automatically the same thing.
Research Sources
- Haines, K.L., et al. “Child Firearm-Related Homicide and Suicide by State Legislation in the US (2009–2020).” Journal of the American College of Surgeons, Vol. 239, No. 6, 2024, pp. 576–587. DOI: 10.1097/XCS.0000000000001144.
- Duke Health. “State Gun Laws Have Mixed Impact on Suicide and Homicide Rates.” July 9, 2024.
- Centers for Disease Control and Prevention. “Firearm Deaths Grow, Disparities Widen.” CDC Vital Signs.
- Centers for Disease Control and Prevention. Research on youth firearm violence, community disadvantage and violence-related risk factors.
- RAND Corporation. The Science of Gun Policy: A Critical Synthesis of Research Evidence on the Effects of Gun Policies in the United States, Fifth Edition, 2026.
- Braga, A.A., et al. “Focused Deterrence Strategies Effects on Crime: A Systematic Review.” Campbell Systematic Reviews.
