Violence followed the soldiers of the Fourth Brigade, First Armored Division, when they returned from Iraq to Fort Bliss in El Paso, Texas, in March 2008. Four soldiers from the brigade took their own lives that year. It was more than had died in combat on their last deployment. “Being back in garrison is what we don’t do well,” Lt. Col. David Wilson said at the time.

The suicide rate in the military used to be lower than the population at large. But in the years following the start of the wars in Afghanistan and Iraq, service members began taking their own lives in increasing numbers. The suicide rate for the Army, the branch most affected, peaked in 2012 at 29.9 suicides per 100,000 people, more than twice the national population’s 12.6 suicides per 100,000. The service saw 165 suicides that year, about one death every 2.2 days. Since then, the Army’s rate seems to have stabilized at about 120 suicides per year — still much higher than before the numbers started rising.

For the last five months, I’ve been reporting a profile of Major General Dana Pittard, now retired, who took command at Fort Bliss in July 2010, shortly after the Fourth Brigade returned from another tour of duty. In my story, published Friday by The Trace and Politico Magazine, I describe how Pittard upended traditional military protocol for how best to prevent suicides, making resources available around the clock for all soldiers who might be experiencing depression, and greatly expanding mental health counseling on the base. Suicides on the base declined, and Pittard won attention from top Army brass.

Despite those results, the Army has not sought to replicate Pittard’s reforms, for reasons that appear to have much to do with a divide over how best to find and treat soldiers who might be most at risk.

The Army has favored programs that seek to identify high-risk individuals and either keep them from enlisting in the first place, or give those already in the service specialized treatment. It is an approach that many, including Pittard, believe helps maintain a false perception that suicidal ideation is extremely abnormal, instead of something that can afflict many people, depending on their circumstances.

The Army also doesn’t yet understand what caused the post-wars surge in suicides in the first place, despite millions of research dollars spent in an effort to find the answer. Several theories for the uptick in deaths — that the Army was accepting soldiers with more serious mental health issues than it had before, or that multiple deployments in long-running wars led to more trauma and suicide — have been disproved. In an especially surprising result for many people looking at this phenomenon, a comprehensive study released in May 2016 found that soldiers who have never deployed account for the majority of suicides.

When service members leave the military, they can be even more vulnerable. Twenty veterans die every day from suicide. Guns are used in a relatively small percentage of suicide attempts, but are responsible for two-thirds of all deaths. Despite the risk, the House of Representatives on Thursday approved a bill that would open the door for gun ownership rights to be restored to tens of thousands of veterans who are currently barred for mental health reasons. At the moment, anyone deemed mentally incompetent by the Veterans Administration may not legally possess a gun. The new law would require a judicial finding of incompetency in order to restrict ownership rights.

Figuring out why people take their own lives is a problem that extends far beyond the military. The reasons for suicide are complex, typically multifactorial, and only loosely understood. People can be at high risk because of demographics, because of genetic variations that affect how they respond to emotional trauma, because of a wide range of stressors, life circumstances, and mental illnesses, or because of access to firearms, which are by far the most lethal means. In the military, many of these factors combine.

Both the percentage of suicide attempts and suicide deaths by gun are higher in the military than in the civilian world, possibly because of widespread access to firearms in the armed services. Strategies to reduce access to firearms because of their lethality are known as means restriction, and they have shown some significant results. A much-cited Israeli study examined the impact of a policy change that stopped soldiers from bringing their rifles home on weekends and found a 40 percent decrease in suicides.

The demographics of the military are especially vulnerable to suicide. White men comprise more than half of the active duty military. They are much more likely than others to end their own lives, accounting for 7 out of 10 suicides nationally. What makes the growth of suicides surprising, though, is that the population of the Army hasn’t changed dramatically since the early 2000s. So experts tend to look for psychological factors rather than demographic ones to explain the rising rate.

Jason Roncoroni, a lieutenant colonel who founded a suicide-prevention organization after his retirement from the Army in 2015, attributes the military’s suicide rate to the stress generated by a sense among soldiers that current American wars will never end, and the expectation of endless future deployments.

“We’ve blurred the line between peacetime and wartime,” Roncoroni told me. “So you have this heightened sense of anticipation of what’s to come, which exists as a cloud of anxiety that follows you around.”

Other experts broadly agree. Dr. Elspeth Ritchie, a retired colonel who ran mental health services for the Army Surgeon General, said that units with what’s called high operational tempo — essentially, a high frequency of deployment — tend to have high suicide rates, even if some of the individuals who end their own lives in the unit haven’t personally seen much action yet.

So what’s to be done? The military has implemented dozens of intervention training and support programs to try to address the crisis, notably the ASIST (Applied Suicide Intervention Skills Training) program and something called ACE, which stands for Ask, Care and Escort — the recommended procedure for soldiers to follow when they suspect that one of their fellows may be in danger. The Army has also expanded access to psychological care. Many installations have installed a system called “embedded behavioral health” that places mental health providers and counselors directly in units, with the hope that the proximity will make soldiers more comfortable going to them for help.

All these efforts are showing some signs of success — the rate has come down from its 2012 high. But observers worry that 120 suicides per year has become a new normal for the Army. After all this time, the Army and the military at large still don’t know why the rate is so high, and they don’t know what can bring it back down.