In 2024, nearly 49,000 Americans died by suicide. One of the most devastating aspects of this crisis is that four out of five of them had seen a healthcare provider in the year before their death. These deaths are preventable. Researchers have found evidence-based suicide prevention practices that reduce risk dramatically. What’s missing is implementation at scale.

Implementing these practices is genuinely difficult. Clinicians face real constraints: limited time, little training in suicide-specific care, and concern about liability. But technology is changing the equation, making it possible to embed evidence-based suicide care into existing workflows without overburdening providers.

The Four Pines Fund recently released a research brief clarifying what effective suicide care looks like, and how wide the gap between evidence and practice really is. In this blog we’ll explore those best practices and how technology can help providers efficiently implement them.

The Scale of the Crisis

The brief highlights that suicide rates in the United States have increased nearly 40% since 2000. In 2024, 16.9 million people reported serious thoughts of suicide, and 2.9 million made attempts. For Americans between the ages of 10 and 34, suicide is the second leading cause of death. And the loss extends far beyond just the individual: an estimated 135 family members, friends, and colleagues are affected by every single suicide.

These numbers are staggering, but it’s a problem hiding in plain sight. 

What Effective Suicide Prevention Actually Looks Like

The Four Pines Fund brief organizes effective suicide care into three steps: identify, engage, and treat.

Identify. Universal screening works, and asking about suicide does not increase risk. The Columbia-Suicide Severity Rating Scale (C-SSRS) and Ask Suicide-Screening Questions (ASQ) are validated tools any provider can use. Critically, many people at risk have no prior mental health diagnosis, which means screening only “high-risk” populations misses most of them. Screening also need not be confined to the clinical encounter: in a first-of-its-kind peer-reviewed implementation study, NeuroFlow’s research team found that remote, asynchronous C-SSRS screening delivered via email and SMS surfaced risk that appointment-based care alone would have missed—one clinical director described a patient who had never disclosed suicidal thoughts in session but endorsed risk through remote screening. Individuals often respond more honestly when completing screenings privately, and across diverse care settings, no participating organization reported a harmful outcome associated with remote screening.

Engage. Hospitalization is frequently overused as a response to suicidal ideation and produces limited benefit when suicidal thoughts go unaddressed during the stay. More effective clinical approaches include the Stanley-Brown Safety Planning Intervention, lethal means counseling, and caring contacts–brief supportive outreach that randomized trials have shown to reduce attempts among high-risk individuals. 

Treat. Suicide-specific therapies like CAMS, DBT, and CT-SP are meaningfully different from general mental health treatment. Treating depression alone is not the same as treating suicidality.

The results from organizations that have implemented this framework are striking. Primary care settings in Washington State saw a 25% decline in suicide attempts. Outpatient clinics in New York had up to 70% fewer attempts and deaths over one year. A major U.S. health system saw a 75% drop in suicide deaths over four years. These findings come from separate published studies, all cited in the Four Pines brief.

The Barriers to Better Suicide Care & Prevention

Despite clear evidence, few providers follow these guidelines. Only 4% of hospitals fully implement the safe discharge practices required to reduce suicide risk (Chitavi et al). Fewer than half of patients leaving psychiatric hospitalization receive timely follow-up care. Most clinicians have never been formally trained in suicide-specific care; it is rarely included in medical school curricula or continuing education requirements. New federal billing codes introduced in 2025 for evidence-based practices like safety planning are a meaningful step forward, but adoption remains far behind where it needs to be.

The deeper problem is that most organizations simply can’t see who needs help until it’s too late. Claims data tells only part of the story, and by the time a claim is filed, a person’s condition often already escalated to the point of crisis. Meanwhile, the people most at risk are frequently invisible to the system: many suffer in silence, struggle to seek help, or don’t recognize what they’re experiencing as depression or anxiety in the first place. When they do reach out, they hit long waits for limited behavioral health services.

On the provider side, the constraints compound the problem. Clinicians are stretched thin, with little time to add screening to already-full visits, and many hesitate to ask about suicide at all out of fear of malpractice liability—even though the evidence runs the other way, since a documented measurement and prevention process protects providers, while the absence of one after repeated warning signs is what actually exposes them. 

The result is a reactive system that waits for crises rather than catching risk upstream, precisely because the tools to identify rising risk and the capacity to act on it have not been built into routine workflows.

NeuroFlow Helps Health Systems Provide Suicide Care at Scale

NeuroFlow was built to solve exactly the problems described above. The platform, paired with a team of trained crisis professionals called Response Services, streamlines the three steps that effective suicide care requires: identifying risk across a population, engaging the people who screen positive, and navigating them to appropriate treatment. What once depended on a clinician having enough time, training, and visibility to catch risk in a single appointment becomes a continuous, technology-supported process that does not rest on any one encounter.

When the platform surfaces a potential risk, a team of mental health professionals trained in crisis intervention and supervised by licensed clinicians evaluates the alert and reaches out to the individual directly. The technology handles identification at scale. Response Services provides the human connection, assessment, and follow-through that turn a flag into care.

Identify. NeuroFlow screens on a regular cadence rather than as a one-time intake step, catching rising risk between appointments. But screenings alone miss two groups: people who have fallen out of assessment compliance, and people who complete an assessment without indicating any suicidal thoughts. To reach them, NeuroFlow also analyzes free-text entries like journal entries within our patient-facing platform using Natural Language Processing to detect language associated with elevated risk. In a published retrospective study, this approach surfaced suicidal ideation among patients who did endorse suicidal ideation on the PHQ-9 as well as those who completed a PHQ-9 and did not endorse suicidal ideation and those who were out of compliance with the assessment. Suicidality does not always surface in surveys; people who underreport in a questionnaire will sometimes express it freely in writing. Every flag triggers immediate automated crisis resources, a human review, and, where needed, a Response Services outreach call.

Engage. Response Services reaches out to every urgent alert with a median response time of 14 minutes. Every individual receives crisis resources regardless of whether they connect with a coordinator – a form of caring contacts, the brief supportive outreach that randomized trials have shown reduces attempts among high-risk individuals. When someone does connect, coordinators follow the research backed effective protocol: conduct a structured screener, collaborate on a Stanley-Brown Safety Planning Intervention, and provide lethal means counseling to reduce access to methods of self-harm. They also handle the referral management that connects that person to the right level of care, from behavioral health and psychiatry referrals to community resources addressing financial strain, food insecurity, or social isolation.

When our system recently flagged a high-risk individual, a Response Services coordinator reached out within minutes. The call didn’t look like what most people picture: he was upbeat, even joking, the kind of person the system routinely misses. A brief ASQ screen flagged the need for a deeper assessment, giving the coordinator the full picture before any next steps were discussed. He had recently developed a plan to end his life, but the assessment confirmed he was safe at the moment and that robust support was already in place: an existing safety plan, individual therapy, group therapy, and psychiatry. He willingly agreed to reduce access to his medication, to see his PCP that day, and to contact his therapist for a sooner appointment. No emergency visit. No costly hospitalization.

Treat. Response Services connects individuals to therapists trained specifically in CAMS and DBT. NeuroFlow also performs warm handoffs to referred providers, closing the referral loop that traditional behavioral health referral processes leave open and increasing the likelihood that individuals attend their first appointment.

NeuroFlow analyzed our Response Services data since 2023 and the impact was startling. Across the populations NeuroFlow supports, individuals who engage with Response Services show 26% greater improvement in PHQ-9 scores compared to those who do not connect, and more than 58% of engaged users achieve zero suicidal ideation at follow-up. Safety planning is associated with a projected 43% reduction in suicide attempts. These are the outcomes that evidence-based practices, implemented at scale, can produce.  

For health systems, that kind of upstream prevention also reduces real financial exposure: a single suicide sentinel event can cost an organization anywhere from $1 million to more than $10 million in litigation, investigation, and regulatory response, alongside lasting reputational and emotional toll.

The research is not new. The practices are not experimental.  The missing piece has always been reliable implementation at scale — and that’s what NeuroFlow is built for. 

If you’re a health system or health plan leader evaluating your suicide prevention and referral management infrastructure, NeuroFlow was built to close these gaps. 

We would welcome a conversation.

Related posts:

NeuroFlow’s Journey Gallery Improves Access to Self-Guided Behavioral Health Support Behavioral Health for Veterans: A Better Path Forward How to Address the Veteran Suicide Crisis: Fight the Stigma & Detect Risk Earlier NEHI Advisory Group Explores How to Accelerate Behavioral Health Integration in Primary Care