Jeremy Ramirez is a decorated Army medic who served three tours of duty in the hottest combat zones of Iraq. In one battle, a sniper's bullet shattered a window in his vehicle, and splinters of glass exploded into his face. But Ramirez counts himself lucky - two of his comrades tragically perished that day.
Twenty years later, Ramirez has earned two master's degrees from UC Berkeley and now serves on the health care management faculty at California State University, Long Beach, and he remains deeply committed to veterans' health. With Berkeley data scientists Michael Rivera and D. Alex Hughes, he's leading a study that aims to improve veteran suicide prevention efforts in California and nationwide.
Experts widely agree that suicide among veterans is a long-running crisis. But, the researchers say, California doesn't have an accurate count of veterans who die by suicide every year, and without that basic information, the state's efforts are bound to be inefficient.
Veteran suicides are an "epidemic," Ramirez said. "But we can't fix a problem if we don't fully understand it. If we don't know the magnitude of the problem, how can we effectively allocate resources across California to better address the issues?"
Added Hughes: "I think we have a responsibility, as a society, to understand what is happening to former members of the armed services. … We have a moral obligation to support them."
The research taken up by Ramirez, Rivera, Hughes and their team is far more complex than it seems. But if they can find a way to make a more accurate count of veteran suicides, that could help California and other states expand and focus support, helping to save lives among millions of Americans who are often forgotten after their military service.
A commitment to comrades in arms
Ramirez enlisted in the Army when he was 17, and after boot camp, he trained to be an Army medic. He was assigned to the elite 82nd Airborne Division, which is often deployed to high-intensity combat missions.
A consequence of military service and warfighting is that veterans have a fundamentally different health profile.
D. Alex Hughes
During three tours in the Iraq War, he often was part of the response team for suicide bombings. After one attack, he was the first medic on the scene, providing emergency care to at least 30 victims and helping to evacuate them. For that action, he was awarded the Army's Combat Medical Badge.
Later, after surviving the sniper attack, he was awarded a Purple Heart. But the attack had a deep effect, not just because of his own wounds, but also because of the psychological wound of seeing Staff Sergeant Zachary Tomczak and Sergeant Richard Ford die that day. Even now, he carries their memories as an inspiration.
"If you have the privilege of coming home when others did not," Ramirez said, "you have to make the best of this. Anything else would be an offense to those who did not get that opportunity."
Back in the U.S., Ramirez became an Army critical care nurse at the famed Walter Reed Military Medical Center, then an administrator there. After receiving Berkeley master's degrees in public health and data science, he now is a professor of health information technology at Cal State Long Beach.
Through the California Department of Veterans Services, he learned about the dearth of basic, reliable data on suicide among veterans. He enlisted a team of Long Beach colleagues and brought on Rivera and Hughes, his former professors at the Berkeley School of Information, to investigate the issue.
"Every one of us would describe ourselves as humanists," Hughes said. "My Ph.D. is in political science, but I'm not a political scientist because I care about politics. I'm a political scientist because the consequences of politics affect people's lives. And so when we saw this problem, it just struck us as being deeply important."
Shocking clues, elusive data
Veterans return from service with important social and health needs, and they're often more susceptible to self-harm, the researchers said.
U.S. Army photo
"A consequence of military service and warfighting is that veterans have a fundamentally different health profile," Hughes said. "And they have a fundamentally altered social network upon returning from deployed service or active duty. Reintegrating into civilian life is profoundly challenging for a number of former service members. That fact, in and of itself, is elevating the suicide risk for some individuals."
The numbers may be imperfect, but they are profoundly troubling.
The 2020 U.S. census counted slightly more than 1.2 million veterans living in California, about 4% of the state's population. The 2024 California Veterans Service Agency Report identified 6,135 veteran deaths due to suicide nationwide in 2023, a rate of 33.9 per 100,000 residents.
But the report found that the rate among California veterans that year was 20% higher - 570 deaths, a rate of 41 suicides per 100,000 veterans.
Transgender veterans and active-duty military personnel face the highest odds of suicidality, the report said.
Nationwide, Ramirez said, the count of veteran suicides ranges from 17 to 44 per day. "No matter how you spin this," he said, "that's at least two to three times the rate compared to the general adult population."
Among experts in veterans' health, there's widespread belief that official numbers undercount suicides, possibly by large margin. But nobody knows for sure.
A mix of science and diplomacy
The researchers learned early that there is no unified system for counting suicides among veterans. Federal, state and local agencies have different methods. Each county has a different system; urban counties and rural counties often have far different resources and expertise, which often leads to different approaches in death investigations.
"Every county's death investigator is doing their earnest, best work, but counties just aren't aligned," Hughes explained. "We have asked death investigators whether it is possible that veteran suicides are undercounted, and several have said, 'I guarantee these are undercounted.'"
How could the researchers understand all the variables, and all the different practices? Obviously, this was a problem for data science. But there was also a diplomatic challenge.
To advance the research, they had to expand relationships within the California Department of Veterans Affairs, which funds their research, and with veterans' groups statewide. They assembled an 18-member advisory board, about two-thirds of them veterans.
Because the problem is so complex, members of the research team over the past year have visited each of California's 58 counties, logging thousands of miles on the road. They've talked to law enforcement agencies and medical examiners, veterans' organizations, tribal officials and funeral directors.
Some of their hosts were surprised that researchers from Long Beach and Berkeley would cross the state to meet them. Officials in a few counties were reluctant, Hughes added, but they eventually agreed to work with the researchers.
"The good news is that we were met with a very positive reception," Ramirez said.
Whose job is it?
The bad news is that the reserachers discovered myriad places in the systems of tracking veterans and deaths where veteran suicides might simply fall through the cracks.
In simplest terms, the researchers say, the problem is both human and technological.
Formally, no single office has the job of assessing whether a suicide victim is a veteran. Death investigations are conducted by county offices - sheriff-coroners and medical examiners - but these officials are most concerned with the cause and manner of death.
In many cases, the researchers said, it's the funeral director who learns that the victim was a veteran. But, Hughes said, sometimes funeral directors cannot enter this information into California's Electronic Death Registration System, and they have to circle back to the coroner to update the record. But people are busy, and sometimes the information can slip through the cracks.
The challenge is compounded because counties and other local agencies use different software for tracking death data, the researchers said.
A further problem is that there's confusion about who qualifies as a veteran. Only those who served in battle? Members of the National Guard? The spouses of soldiers?
Hughes suggested that medical examiners, funeral directors and others sometimes have an implicit bias about who's likely to be a member of the U.S. armed forces. They may envision a grey-haired Vietnam vet at the VFW, or a soldier who served in Iraq or Afghanistan, but not a cyberwarrior, for example.
"That can be an inaccurate narrative," he explained. "There are a huge number of female and non-binary veterans. There are a number of non-citizen veterans. So if people who investigate deaths are looking for veteran status, there's a possibility that they look with more fervor when the deceased is male rather than female, or a citizen rather than non-citizen."
California can be a model
Ramirez, Rivera, Hughes and their colleagues are still working on the study, and they haven't yet developed their recommendations. But, they say, identifying gaps and kinks in the information system suggests possible solutions.
Improved technology is one. Formalizing the responsibility for identifying veterans who have died by suicide or other causes is another. It may be just as important to expand awareness among medical examiners, funeral directors and others.
Ideally, the researchers say, the California Department of Veteran Affairs could use their study to help develop better, more targeted suicide prevention and other support services. Other states might then follow California's lead, with a goal of improving veterans' lives - and saving lives - nationwide.
"We think every state in the country should be doing something similar," Ramirez said. "We want to be able to help out as best as we can."
