The 34th Infantry Division of the Minnesota Army National Guard has deployed more frequently since 9/11 than almost any other unit in America. In 2011, 2,700 soldiers from its 1st Heavy Brigade Combat Team were sent to the Middle East to assist with the drawdown in Iraq. One of its battalions spent that summer at Wisconsin’s Fort McCoy, preparing for deployment under Lt. Col. Charles Kemper. Before they boarded a plane for Kuwait, however, Kemper led his troops in a brief ceremony.

“He had every single non-leader get down on their knee and look their leader in the eye and say, ‘I promise not to kill myself.’ And then all the leaders got down on their knee and told their soldiers, looked them in the eye, and said, ‘I promise not to kill myself,’” recalled Sgt. 1st Class Joshua Guyse, an infantryman who was then part of the battalion’s Bravo Company. “I’ve never experienced … I mean, the whole battalion. I don’t know how many guys it was in total. Four, five hundred, standing out there in a field. I’ve never experienced that before in my military career. He got it. That colonel. He understood.”

That is to say, Kemper understood how to reach soldiers with a message of suicide prevention. Guyse, who narrowly averted his own suicide attempt in 2009, told me this ceremony held a deep meaning that goes beyond simply repeating a phrase. A promise made to the people whose lives you hold in your hands and who, in turn, hold yours, is a blood pact that cannot be easily broken.

Sgt. 1st Class Joshua Guyse in his living room in Swanville, Minnesota.

Sgt. 1st Class Joshua Guyse at his home in Swanville, Minnesota.

The United States Department of Defense has acknowledged that it has a problem with suicide. Today, it’s one of the leading causes of death in the military — and that’s without including suicides among veterans. For generations, though, the military enjoyed a remarkably low suicide rate — lower than that of the general population — nobody knows why. But that began to change after Sept. 11, 2001. That year, 160 active-duty members of the Army, Navy, Air Force and Marines killed themselves. The combined suicide rate was 10.3 people per 100,000. Today, the Department of Defense publishes information on those four main active branches in the annual DOD Suicide Event Report. In 2014, the latest year with complete data, 269 troops killed themselves, and the total active component suicide rate was 19.9. The Army National Guard, which is counted separately, had 89 suicides and a rate of 21.5. In contrast, the overall American suicide rate in 2014 was 13.4 per 100,000. Just like nobody knows why the military suicide rate used to be low, nobody knows why it went up, either.

The public explanation for this increase has focused on the uniqueness of the military experience, specifically the wars in Iraq and Afghanistan and combat-related post-traumatic stress disorder. The military’s rising suicide rate is treated as specific to the military, every bit as tailored to this population as Kemper’s suicide-prevention technique. But this narrative is frustrating to the scientists who study military suicide and to the troops who live with the military experience, who say it is misleading.

Instead, they think it’s important to note that the overall American suicide rate has also risen in the past 16 years. Suicide is the 10th leading cause of death in America. It makes up the largest category of gun deaths in this country, with 21,334 deaths in 2014. In both the military and the U.S. as a whole, guns are used in at least half of all suicides. And in the military, the raw number of suicides is small. Consider the 269 active-duty members of the Army, Navy, Air Force and Marines who killed themselves in 2014. That’s 269 too many, but it’s a small-enough number in a population of more than 1 million to make statistical analysis tricky.

The comparison I made earlier between the DOD’s suicide rate and that of the nation as a whole is also misleading. That’s because the military consists largely of young white males — a segment of the population in which suicide is particularly common. If you adjust the U.S. suicide rate to account for these demographic differences, the disparities between the national rate and the military rate blur and fade, said Craig Bryan, assistant professor of clinical psychology at the University of Utah. He’s one of the scientists who have calculated what the estimated suicide rate would be for the U.S. if the nation had the same demographic breakdown as the military. Doing this, comparing apples to apples, produces an American suicide rate of about 20 per 100,000 people, or just about the 19.9 rate we see in the military.




What’s more, research shows that deployment and combat are not the best predictors of suicidality in the military. In general, the military’s suicide problem looks a great deal like America’s suicide problem. The same people, the same risk predictors and the same means.

“The military is both a part of and apart from the rest of the nation,” said Dr. Robert Ursano, a civilian professor of psychiatry and neuroscience at the Uniformed Services University in Bethesda, Maryland. “The things that happen in the nation and what the nation represents, we see in the military.”

Ursano, Bryan and other scientists say that matters, because it affects the warning signs that family and friends of soldiers should look out for — not only people who have gone to war are at risk. The similarities also mean that efforts to understand, prevent and treat suicidality in the military can aid our efforts to do the same in the civilian world — and vice versa.

A county road not far from Guyse's house. A view of the waiting area of the VA hospital in Minneapolis, where Guyse was treated for depression and PTSD.

A county road near Guyse’s house. The waiting area of the Minneapolis VA hospital where Guyse was treated for depression and PTSD.

In May of 2009, Joshua Guyse climbed into a car in Rochester, Minnesota, with his aunt and uncle, with whom he’d been living. He thought they were taking him house hunting. Instead, they’d planned an intervention. They drove Guyse to the Veterans Affairs Hospital in Minneapolis and checked him into the emergency room. He still isn’t sure what tipped them off that he was in trouble. He looked perfectly healthy. But internally, he was deep into a depression that was close to taking his life. He had his suicide all planned out, he told me. That kind of emergency is every bit as ER-worthy as a broken leg or a bleeding vein. “If somebody has suicidal ideation you have to get them in right away, and trying to schedule a clinic visit … that may take weeks. The safest route is to take somebody to the ER,” he said.

The road that took Guyse to the emergency room door began on Sept. 11. Guyse grew up in the San Francisco Bay Area. His middle name is Rain. His parents were, shall we say, less than thrilled with his choice to join the Army out of high school in 1989. But his service was uneventful, and he never deployed. He was honorably discharged in 1999. And then 9/11 happened. “I kind of made a decision in life: Either you’re part of the problem, or you’re part of the solution,” he said. “All politics aside, the global war on terrorism was something I truly believed in.”

So, a few years later, Guyse re-enlisted, this time with the Army National Guard, which he chose because it gave him a chance to serve in a combat position, but also to live where he wanted. He deployed to Iraq with the California National Guard for 15 months and came home in September 2007. Over the next year and a half, his life fell apart.

Guyse had encountered multiple struggles during his time in Iraq. First, the culture shock that anybody would experience moving halfway around the world. Then there was the uncertainty of never quite knowing whether, say, that guy and his goat driving by in a pickup were a normal part of human social variation or an incoming threat. His unit took some casualties. He didn’t understand some of the choices his leadership made and had no power to question or control them. There was never any big, dramatic event for him in Iraq. But, over time, he just got more aggressive, more angry, more numb. “Mom, God, country, apple pie — all that stuff, it just has no meaning to me anymore. You go in with this grandiose idea: I’m going to go in and slay the big, evil dragon, and then there just isn’t. You try to process things the way you used to, and you can’t.”

His experience in Iraq certainly primed him for trouble. But it was compounded by much more prosaic life problems after he got home: His marriage fell apart, he had trouble finding a job, he grieved the loss of his community of soldiers as he tried to slip back into civilian life. He packed up and moved to a new state far away from many of his family members and friends. And all of those things — those are suicide risk factors, too.

Guyse's pack and personal protective equipment in his home office. Guyse in Iraq.

Guyse’s pack and personal protective equipment in his home office. Guyse in Iraq.

Iraq photo courtesy of Joshua Guyse

This is the pattern that researchers such as Craig Bryan are finding as they dig into the demographics and risk factors associated with military suicide. “Even among those with combat PTSD, we’re overwhelmingly seeing that the factors leading service members to suicide are not combat-specific,” Bryan said. “I think the best way to think about it is that PTSD can lower a person’s threshold for tolerating emotional pain and distress, and disrupts many other aspects of their lives.” But PTSD alone doesn’t seem to be causing suicide. Instead, “when a major stressor like a relationship problem occurs, the veteran with PTSD ends up having less ability to tolerate that stress.”

He’s not the only one to find this. In 2013, researchers associated with the Naval Health Research Center analyzed military suicide data on troops from the four branches plus Coast Guard, National Guard and Reserve members, which broke down the suicide rate by factors including whether the victim had deployed. They found that deployment and combat exposure did raise the suicide rate, but not by a lot. Troops who had never deployed had a suicide rate of 10.7 per 100,000 person-years,1 those who deployed with no combat exposure had a suicide rate of 12.8, and those who saw combat had a suicide rate of 12.4. Yes, the people who saw combat had a slightly lower rate than those who did not.

That’s not the only surprising finding: Combat specialists, the soldiers who actually shoot other people, didn’t have the highest suicide rate. Mechanical and electrical repair staff did. Nobody thinks combat is causally protective against suicide, just like nobody really thinks there’s something deeply traumatic about being a mechanic. But this data suggests that there is more contributing to military suicides than simply whether someone has fought a battle.

Patterns of suicide in the Army National Guard suggest something similar. The Guard has been pulling out this data and publishing it since only 2007, and, in the intervening years, its suicide rate has at times exceeded that of the demographic-adjusted general population, the active-duty Army and the DOD as a whole. More than 56 percent of the 294 Guard suicides from 2007 through 2010 were among people who had never deployed. The same paper also found that post-deployment stressors such as the end of a relationship, job difficulties and major life changes were predictive of suicidal thoughts — and combat wasn’t. Overall, the biggest predictors were age (25 and younger), sex (male) and race (white).

A metal plaque of the U.S. Army seal hangs on the wall of Guyse's living room. Guyse stops to chat with soldiers at Fort Ripley, where he works as an infantry instructor.

A plaque of the U.S. Army seal on Guyse’s living room wall. Guyse, in blue, chats with soldiers at Camp Ripley in Minnesota, where he is an infantry instructor.

The use of guns as the means of suicide is also a major factor connecting suicide in the military with suicide in the civilian world. The primary difference is that military members who take their own lives are likelier to use a gun, and women in the military are significantly more likely to use a gun (and, relatedly, more likely to complete a suicide attempt) than civilian women.

This tracks with a popular theory of suicide that frames the act as somewhat impulsive, according to Carl Castro, an assistant professor and director of the University of Southern California’s Center for Innovation and Research on Veterans and Military Families. Suicide is impulsive not in the sense that it happens out of nowhere — people who attempt suicide have usually been planning it for a while, just like Joshua Guyse was. But the shift from plan to action is often a spur-of-the-moment decision, Castro said.

When they decide to act, suicidal people reach for tools that are familiar, comfortable and easy to access. For instance, a study that looked at military suicides in basic training from 1977 through 2001 found that recruits tended to choose methods that matched their skills and experiences — Army recruits used their military-issued M16 rifles, while Navy and Air Force recruits (who had less access to weapons) tended to die by hanging and falling/jumping, respectively. Marines died by all three methods.

In this theory, more military women kill themselves with guns because military women are more comfortable and familiar with guns than civilian women tend to be. Likewise, military members, as a whole, die by suicide by firearm at a higher rate than civilians not because their work requires them to have guns — most soldiers don’t take their military-issued weapons home, the guns used in military suicides aren’t usually military-issued, and most military suicides happen in settings that aren’t specifically military — but because people who join the military tend to be people who like firearms, are comfortable with them, and own some for personal use.

Guyse keeps several guns in a large safe in his home. This one is a Japanese training rifle that fires blanks.

Guyse keeps several guns in a safe in his home. This one is a Japanese training rifle that fires blanks.

Joshua Guyse’s suicide plan in Minnesota did not involve a firearm. He’s not really sure why. It probably had a lot to do with the fact that, at the time, his personal guns were in his brother’s safe, back in California. That’s exactly the kind of thing public health experts wish more suicidal people would do with their personal firearms — temporarily turn them over to a trusted friend or family member, the same way you’d turn the keys over to your designated driver after a night at the bar.

There is reason to think that means-restriction techniques like this could reduce the suicide rate for both military and civilian populations. That’s because guns are a particularly lethal way to attempt suicide. In 2006, the Israeli Defense Forces tried this, changing policies that had once allowed soldiers to take weapons home with them on weekend leave. Over the next two years, the suicide rate dropped by 40 percent — almost all of that accounted for by a reduction in weekend suicides. Weekday rates stayed pretty stable.

The U.S. Department of Defense has implemented programs that are aimed at approximating this kind of means restriction. The Army National Guard, for instance, has passed out more than 10,000 gun locks in 2016 alone. The locks come from a VA program and have a suicide help line number printed on them; the idea is that the 10 minutes it takes to fumble with a lock is 10 minutes you have to change your mind or seek help. But Maj. Carmel “Clint” Osborne, Army National Guard suicide prevention section chief, said there wasn’t a good way to track what happened to those locks or whether distributing them reduced the number of suicide attempts.

The main entrance to the VA hospital in Minneapolis. The complex where Guyse lived while he was being treated.

The main entrance to the VA hospital in Minneapolis. The complex where Guyse lived while he was being treated.

Compared with other suicide-prevention efforts, much of the military’s means-restriction work feels a bit ad hoc. It’s happening. But it’s not as widely promoted as other suicide-prevention programs, and most of it falls into the category of “things we encourage people to do” rather than “things we are making people do.”

In 2011, the Rand Corp. published what James Griffith, a retired Army colonel who is now a research psychologist at the University of Utah’s National Center for Veterans Studies, called the first major report on suicide in the military. At the time, only the Air Force had any policies relating to means restriction. Almost all the efforts were focused on awareness, self-care and identifying people who might be at risk. That really hasn’t changed. As a Congressional Research Service report on suicide prevention in the Veterans Health Administration put it, “Means restriction is a promising area still under investigation.”

The Rand paper concluded that means restriction was possible in the military context, Griffith told me. But, he said, “I don’t think that’s gone hardly anywhere. When I say it, they look at me like I’m talking gun control.”

From Joshua Guyse’s perspective, this isn’t a problem. “You can talk about gun safety, but infantry guys are around guns so much, they know what’s safe and what isn’t,” he said. As for asking people to turn over their weapons, even if it’s just to a friend, Guyse worries that could hurt suicide-prevention efforts, rather than help. Suicidal ideation — and, really, any kind of mental health problem — still carries a lot of stigma in the military. The main reason Guyse is so open about his experiences is because he wants soldiers to see that you can still have a career after being hospitalized as a suicide risk. He’s been promoted since his treatment and is now a senior infantry instructor with the 2nd Battalion, 175th Regiment, stationed at Minnesota’s Camp Ripley Training Center. He even deployed to Iraq a second time, in 2011, and had no relapse in his suicidal ideation. In his civilian life, he recently graduated from nursing school.

But the stigma already deters soldiers who feel suicidal from bringing their problems to their commanders or even their doctors, Guyse and the psychologists I spoke with all said. If those people also have to worry about their Second Amendment rights and whether they’ll ever be allowed to have their guns back, Guyse thinks more will stay silent.

Guyse’s dog tags from his National Guard deployment to Iraq. Guyse walks down a dirt road that runs from his house to the Swan River.

Guyse’s dog tags from his National Guard deployment to Iraq. Guyse walks down a dirt road that runs from his house to the Swan River.

That gets us to a major problem in addressing military suicide — and, really, all suicide. If we know you are contemplating killing yourself, we have programs that can help you. For instance, dialectical behavioral therapy — a practice that focuses on mindfulness and emotional regulation — is a treatment Craig Bryan swears by. Guyse received it when he was being treated by the VA. He says it helped him so much that he now uses elements of it with the soldiers under his charge — imagine a bunch of infantrymen sitting around, practicing chair meditation. Systematic reviews of the literature suggest that DBT can be effective at reducing self-destructive behaviors, though a lot of research is still needed, and, according to Ronald Kessler, a Harvard sociologist who studies suicide in the Army, there’s unlikely to ever be a one-size-fits-all treatment.

But the bigger issue is that, most of the time, we don’t know who is on the verge of suicide and who isn’t. Guyse’s aunt and uncle saved his life when they loaded him into the car in May 2009, but that’s the sort of decision that’s often very difficult for a friend or family member to make. Whenever a suicide happens in the Minnesota Army National Guard, leadership and soldiers can go back and find little clues everywhere, in retrospect, said Maj. Ronald Jarvi, the Minnesota Guard’s program manager for suicide prevention. But no one person had all the pieces, so nobody figured out what was happening before it was too late.

The Army is trying to make it easier to spot at-risk troops via the Study to Assess Risk and Resilience in Servicemembers — the largest study of suicide ever conducted, in the military or anywhere.

In the meantime, though, much of what the military does is try to inoculate its entire population against suicide, regardless of individual risk. Some of these programs, like the Minnesota Army National Guard’s Beyond the Yellow Ribbon campaign, focus on the community. Beyond the Yellow Ribbon offers support to soldiers and their families before, during and after deployment. It involves a network of Family Assistance Centers, veteran employment initiatives and other activities aimed at reducing the work-money-family sort of suicide triggers before they happen.

Other programs focus on individual coping skills. The Army’s Master Resilience Training is a great example of this — it’s a highly promoted, highly touted suicide-prevention program that every recruit must go through. The Army National Guard, including the Minnesota Guard, uses elements of it, as well. But there’s no independent evidence that it prevents suicide.

Soldiers get gun safety lectures. They get PowerPoints on suicide warning signs. They get awareness events, such as the 5K run the Minnesota National Guard has planned for later this year, which incorporates suicide-prevention information as part of a scavenger-hunt-type activity. Their doctors get trained in spotting depression and suicidality. There are ceremonies like the one Guyse witnessed. Overall, the sense one gets looking at the military’s suicide-prevention response is of a flurry of activity — far more programs and events than any civilian employer would ever throw at its workers — and not a lot of results. All of which underscores just how little is known about suicide, in general.

STARRS, the study, is likely to help that, and not just for the military. Because suicide in the military doesn’t appear to be all that dissimilar to suicide in the general population, what Ursano, Kessler, and the other STARRS researchers learn about it should be applicable elsewhere. It’s producing some useful results already. The VA is about to start using an algorithm, developed from STARRS data, that can spot the people at highest risk of suicide with a high degree of accuracy.

But prevention remains a difficult nut to crack. “It’s not like there’s this program somebody developed and nobody is using it,” said Castro, the USC professor. “There’s not this panacea, where somebody solved the problem and DOD or whatever is just too lazy or cheap to implement it. There is no evidence-based suicide-prevention program.”

This article is part of our project exploring the more than 33,000 annual gun deaths in America and what it would take to bring that number down. Our podcast What’s The Point is highlighting the project all week.


  1. Studies that follow a group of people over time, as this one did, tend to measure incidence in person-years, rather than in number of people. Say you took 10,000 people and kept track of them for 10 years. If, during that time, 25 of them developed a disease, then that would be a rate of 25 per 100,000 person-years.

Maggie Koerth was a senior reporter for FiveThirtyEight.