Upstream suicide prevention research demonstrates importance of looking beyond mental health

Suicide is often thought of as the result of mental health struggles; however, over half of people who die by suicide do not have a known prior mental health condition. Many of these people die on their first suicide attempt, sometimes with little or no indication to those around them that they are considering ending their life.

One thing that distinguishes this population is that they are more likely to die in the context of specific stressful life circumstances, such as relationship, health or financial problems, life stressors and recent or impending crises, than other people with a history of suicide attempt.

Given that 50–60% of people who die by suicide do not tell anyone beforehand, a core challenge of suicide prevention today is how to identify those people most at risk of falling through the cracks and design upstream interventions to help them.

Two different research projects led by the USC Suzanne Dworak-Peck School of Social Work seek to illuminate and address this crucial gap in suicide prevention.

Both focus on people who are experiencing identity wounds due to crisis or moments of high stress and transition in their lives, rather than traditional mental health crises. Both also look to identify key resources within the social and professional networks of these individuals, who may be able to identify early warning signs, but who are outside of the mental health system and usually do not have any training on what to do.

John Blosnich, associate professor and director of the Center for LGBTQ+ Health Equity, is principal investigator on a study that seeks to identify and partner with what he terms “industries of disruption”—businesses that serve a disproportionate number of people experiencing major life stressors who may be at greater risk for suicide. His initial phase focuses on family law attorneys, up to four in 10 of whom he finds have had a client die by suicide.

“Many of the suicide cases we see are a result of bad things happening to good people and them not knowing how to deal with it or where to turn,” Blosnich said.

“Often, they are having a natural reaction to a really bad string of luck. It would be odd to endure some of these things and not be depressed or anxious. The question is how do we identify those people and interrupt those thought patterns that are contributing to someone saying that life is not worth living anymore?”

Eric Rice, professor and co-director of the USC Center for AI in Society, is principal investigator on a study examining military suicide risk using machine learning to predict with whom a servicemember is most likely to share suicidal thoughts.

Findings demonstrate that these are most often informal resources, often family members. Rice and his team are further refining results with a goal of providing paraprofessional gatekeeper training for these informal resources, so that they have the tools to be effective in intervening proactively upstream of a suicide attempt.

“The emerging disruptions in suicide prevention are around informal supports,” Rice said. “We are trying to identify the places and spaces where we know populations are suffering, and identify the people in those spaces to provide the help they’re already doing in a more empowered and effective way.”

Seeing good people during their worst moments

The U.S. National Strategy for Suicide Prevention has named attorneys as a priority population for suicide prevention training. Family law is an area of particular interest, given that approximately 27% of suicides involve intimate partner problems, including divorce.

“Family law has been mentioned in the same breath as suicide prevention for a very long time but really just anecdotally,” Blosnich said. “Despite this profession being so highly exposed to suicide, this is the only research to date to specifically asking attorneys about suicide exposures and what they would like in terms of suicide prevention training.”

Blosnich and his team are conducting a mixed methods study, “Seeing Good People During their Worst Moments: Pioneering Upstream Suicide Prevention for Industries of Disruption,” including an online survey of 294 family law attorneys and interviews with 17 of these attorneys about suicide exposure and suicide prevention for their clients.

The study is part of a larger examination of industries of disruption, including banks, mortgage providers, employment workforce development boards and the self-storage industry.

Results demonstrate that the problem is significant, with 36% of attorneys reporting that they had a client die by suicide, and 70% indicating that they had a client who they believed to be suicidal.

“It’s often said that in family law they see good people at their worst,” Blosnich said. “It is a unique industry that is a natural bottleneck for people who are experiencing a tough time in their life.”

Blosnich tells the story of one family law attorney interviewed as part of the study who drove around in the middle of the night looking for her client because she feared that he was going to kill his soon-to-be ex-spouse, and then kill himself.

“Family law attorneys are just as exposed to losing someone to suicide as people who work in mental health professions,” Blosnich said. “The difference is that these folks have zero mental health training.”

He notes that many of the things that people experience are social determinants of health and that it is important to examine not only their biological, physical and mental health factors, but also the context in which they live.

“The person in context is what really gives rise to the person we see in front of us,” Blosnich said. “From a social work and public health perspective, why wouldn’t we question these contextual factors as potential intervention points before they get to the point of harming themselves?”

For insight on potential solutions, Blosnich and his team are turning to the attorneys themselves and integrating their perspectives with existing suicide prevention expertise.

Participation in this study indicates that the attorneys recognize the problem and want training on how to deal with it. Nearly two in three attorneys surveyed reported openness to attending in-person or web-based suicide prevention training.

Blosnich is currently identifying potential partners to develop and offer gatekeeper intervention programs tailored to the unique needs of family law attorneys, such as attorney-client privilege and the potential for mental health treatment to be weaponized against their clients.

Increasing informal networks for servicemembers to lean on

U.S. soldiers are more likely to die by suicide than any other cause, with suicide rates among veterans measuring 1.5 times higher than the general population.

Many people assume that increased military suicide risk is due to post-traumatic stress disorder (PTSD) or combat-related stress, but research has shown that combat deployment is actually associated with a lower risk of military suicide.

In fact, Rice and other researchers have found that, in many ways, military suicide risk is more similar to civilian suicide risk than it is different. As with civilians, it is often related to stressful life transitions rather than preexisting mental health conditions, which are involved in just 45% of military suicides.

Military suicide risk is at its highest during three critical stress points when servicemembers often experience core identity issues that may shake their foundational sense of self. The first is when they initially join the service, the second is after returning from deployment and have difficulty adjusting, and the third is when they separate from service and transition back into civilian life.

Also similar to civilians, one of the most effective suicide prevention tools for servicemembers and veterans is self-disclosure to a trusted confidante. However, the stigma around suicidal ideation in military culture and the perception of being “weak” mean that, according to Rice, only 28% to 34% of military personnel who die by suicide tell anyone what is happening until it is too late.

“If we’re always waiting for the person who’s in distress to disclose, we’re losing an opportunity to help,” Rice said.

Rice and his colleagues are conducting a study to better understand the specific determinants of disclosure among Army personnel, following them as they prepared for combat deployment.

“Mindfulness and Help-Seeking in Social Networks to Understand Suicidal Ideation” uses machine learning to identify those people within the servicemembers’ networks to whom they are most likely to disclose suicidal thoughts. It is the first study to apply this machine learning methodology to suicide-related disclosure.

The 241 active-duty U.S. Army personnel participating in the study were recruited from a single battalion and completed a self-report questionnaire addressing who they would talk to if thinking about hurting themselves or planning to attempt suicide. They also participated in a social network interview that identified up to 10 people they had interacted with over the previous month, the quality of those relationships and likelihood of suicide-related disclosure.

Machine learning was chosen because traditional statistical models are not as well-equipped to identify and accommodate complex relationships among various predictive behaviors or attributes.

Rice and his team are using decision-tree models, a type of machine learning that is able to handle more complex, robust data to identify distinct subgroups of a population who share a profile of characteristics influencing a specific behavior, such as suicide-related disclosure.

Results have found that the quality and type of relationship was more predictive of a willingness to reach out during challenging circumstances than any factors about the individual servicemember themselves.

Just 6% of respondents said they would consider suicide-related disclosure to someone in their network who had not previously been an emotional or information support. That number jumped to 86% if the person was a current source of both types of support.

This points directly to the need to develop capacity among servicemembers and the individuals in their networks for specific suicide risk and prevention training that could reduce barriers to disclosure.

“This unique multi-level treatment of disclosure could help to identify novel prevention targets among military personnel,” Rice said. “It may also emulate in broader populations to advance the science of suicide prevention.”

Carl Castro, professor and director of the Center for Innovation and Research on Veterans and Military Families at USC Social Work, as well as director for the RAND-USC Epstein Family Foundation Center for Veterans Policy Research, points out that when all of the attention is focused on the mental health domain, it excludes other risk factors that are highly relevant for addressing suicide prevention.

“People assume you have a mental health disorder when you die by suicide,” said Castro, a retired U.S. Army colonel and co-investigator on the study with Rice. “The data just doesn’t support that. Research points to identity, how people see themselves and how they want other people to see them. A lot of people who die by suicide are suffering from what we term ‘moral injury.'”

Rice explains that these types of upstream interventions, which acknowledge all the risk factors, are the new frontier in suicide prevention, supplementing the very effective traditional interventions for those who reach out for help with suicidal ideation or after an attempt. Together, they can create a more complete prevention system.

“Traditional suicide prevention solutions are often more downstream than what John Blosnich and I are looking at,” Rice said. “We are both trying to identify and leverage unusual places and informal resources to be a more engaged part of the lives of people who are in distress.”