A year ago, Democrats in the Michigan state Senate introduced a bill designed to create an extreme risk protection order law, or ERPO, in the state. Intended to help prevent gun suicides and mass shootings, ERPOs are sometimes called “red flag laws.” They’re basically a timeout for gun owners — a system for temporarily removing weapons from people or preventing them from purchasing them, usually for around a year — when they are deemed to be a threat to themselves or others. Introducing an ERPO in Michigan probably seemed like a sure bet.
Widely supported by experts, ERPOs are one of the most popular types of gun laws out there. They’re favored by a majority of Americans, including a majority of Republicans and a majority of gun owners. Today, 19 states have some form of ERPO on the books. But Michigan — the site of America’s latest high-profile mass shooting — is not one of them.
Despite bipartisan appeal that has led to passage in eight states with divided governments and one, Florida, with a Republican-controlled government at the time of passage, ERPOs remain controversial. Opponents in Michigan told local news the laws violated the Bill of Rights. Even many of the states that do have them rarely use them. In the places where ERPOs already exist, experts and authorities told me that they’re learning that getting one passed and making it work requires trust — between gun owners and the government, law enforcement and the community, and researchers and the systems they’re trying to study.
Advocates for ERPOs are hoping these laws can help prevent both suicides and mass shootings — events that often involve legal gun owners (or young people who live with them) who haven’t done anything before the event that would justify arrest or hospitalization but may have developed behavioral changes that seem deeply troubling to family and friends. Tara Sklar, faculty director of the Health Law & Policy Program at the University of Arizona’s law school, became interested in studying ERPOs after she noticed her elderly father, a military veteran and gun owner, developing sporadic bouts of delirium. “When that occurs, you’re just not speaking to the same person that is this wonderful loving good citizen,” she said. “I was increasingly concerned about … himself and those around him. It completely wasn’t addressed in any of the gun laws.”
There is some data that suggests these laws may be helpful. While a systematic review of scientific literature by the Rand Corporation found no studies of ERPOs that met their standards to know whether the laws prevent suicides or mass shootings, those standards excluded studies that other researchers find compelling. For example, a 2017 study of the Connecticut ERPO, the oldest such law in the country, used data from 1999 to 2013 to estimate that every 10 to 11 gun seizure cases prevented one suicide, an estimate that accounts for suicides carried out after the ERPO expired or where those who died by suicide ended up using other means besides guns.
But that same research found the people requesting ERPOs, the law enforcement serving them, and the people they are meant to protect don’t necessarily understand or trust this system. “If you see it as access to treatment … or access to something that’s going to prevent harm, then you might say, well, thank you, I needed that. It helped me,” said Jeffrey Swanson, a professor of psychiatry and behavioral sciences at Duke University School of Medicine and one of the authors of the 2017 study. “[But] if you think of it as coercive and something that’s violating your rights, or stigmatizing, then you might not.”
Nobody knows how subjects of ERPOs feel about it afterwards, but Swanson’s own research on people who have been involuntarily committed to outpatient psychiatric institutions found that only about 27 percent later said it was the right thing to do and beneficial to them. He thinks it’s likely that ERPOs – which also involve involuntarily, temporarily restricting someone’s rights because of a risk they could harm themselves or others – could run into similar issues.
And that’s not the only debate over the best way to make an ERPO work. For example, who should be able to request an ERPO varies widely by state. Based on her experiences with her father and what she’s seen in subsequent research, Sklar told me that she prefers the versions that enable family members or doctors to request an ERPO, because having more people capable of doing that is likely to mean more problems are caught before they rise to the level of police intervention.
But Sheriff Grady Judd, an ERPO advocate whose office administers these orders in Polk County, Florida, prefers a system where only law enforcement can request an ERPO. In fact, he told me that he’s fought against efforts in the Florida legislature to expand the range of people who can file there because he sees that as opening the system up to abuse. “It’s easy for relatives, neighbors, soon-to-be-ex-wives to go down and swear under oath that this person did something,” Judd said. The gun owners I spoke with shared this concern about the possibility of retaliatory or false reporting. (There is no data on the number or share of false reports available.) Judd thought that problem was best avoided by ensuring all reports had to go through law enforcement and an investigation.
ERPOs involve the government taking away someone’s guns, but serving them seems to involve minimal conflict or danger to law enforcement. That has surprised even Dr. Garen Wintemute, an emergency medicine physician and director of the Violence Prevention Research Program at the University of California, Davis, Medical Center. Before studying California’s ERPO laws, he assumed someone would get hurt while serving an order. “I kept saying, as an ER doc and thinking the worst … [The gun owner will] be really ticked off. He’s going to give you all the guns but one and he’s going to kill himself or somebody else. And that hasn’t happened,” he told me. Wintemute is part of a team studying the effects of a California state law, related to ERPOs, that enables law enforcement to remove guns from owners who later break a law or develop a condition that would make them ineligible to own guns. While the data has yet to be published, he said he’s seen thousands of interactions with no one ending up in the hospital, no guns being discharged and no serious violence.
Law enforcement in Maryland and Florida told me they’d experienced similar patterns. “The pleasant surprise is that a majority of these people recognize they’re in the middle of a mental health break, if you will, and they voluntarily give us the firearms. Many of them don’t even show up to contest the temporary forfeiture,” Judd said.
And while Wintemute said the California law he’s studying involves a dedicated team of officers who go to a subject’s house, other law enforcement said they’ve had good luck with lower-key interactions. For example, Ronald Howard, the sheriff of Somerset County, Maryland, told me he starts by calling the subjects of ERPOs and asking them to bring their guns into the sheriff’s office. Most do. In Florida, Judd starts the process by working with the subject of an ERPO and their community to see if there are family members or friends the subject can turn their guns over to, instead of the sheriff’s office.
But much of what works best for implementing ERPOs — and why — remains anecdotal and undocumented. That’s frustrating to Dr. Emmy Betz, director of the Firearm Injury Prevention Initiative at the University of Colorado. In the years since Colorado enacted an ERPO in 2020, she told me there hasn’t been much data collected or shared on what happens when law enforcement retrieves a subject’s guns or how and when subjects are getting their guns back.
These are questions that concern both researchers and ERPO skeptics. Swanson told me there’s little data on the race of those subject to removal orders, but the limited research that exists suggests there’s potential for bias against Black owners.
The big takeaway for states that want new ERPOs is that trust and collaboration need to be part of the process, Betz said. “If it’s the firearm-owning community who’s likely to be using the law, shouldn’t we be talking about their development and implementation with that community?” she said. Doing that can make a difference. For example, Betz has collected data on who is filing ERPOs in Colorado and what kinds of requests local judges are approving and denying, and she’s shared this data with community members like Colorado gun dealer Jacquelyn Clark. Clark told me that Betz’s data has helped her feel more confident that the system isn’t being used as a weapon against gun owners who don’t need it.
This level of trust could become important in Michigan. Unlike last year, the state’s executive and legislative branches are now controlled by Democrats. In the wake of this recent mass shooting, they’ve vowed to put red flag laws back on the table — and now have the power to make sure it gets passed. But if these advocates want an effective ERPO, they’ll need more than a law — they’ll need the confidence of the people the law affects. “Just having a law on the books doesn’t fix anything if it’s not used,” Betz said.
CORRECTION (Feb. 21, 4:15 p.m.): A previous version of this story said that the little data that exists on racial bias in gun removal orders suggests Black owners are overrepresented. The research, which was conducted in a single county, found evidence that Black gun owners were overrepresented in removal orders. It’s not clear that this represents bias in that county or nationally.