Recognizing and Responding to Suicide Risk: What Clinicians and Counselors Need to Know

If you or someone you know is in crisis, call or text 988 to reach the Suicide & Crisis Lifeline, available 24/7. If there’s immediate danger to someone’s life, call 911.

Suicide risk rarely shows up as a single red flag. It’s usually a combination of factors: a history that raises someone’s baseline risk and often an event that activates that risk into something urgent, with warning signs as the clearest evidence that a crisis is close or already underway.

For clinicians and counselors, the core skill is telling the difference between a patient who needs continued monitoring and one who needs immediate action. This piece covers how to make that call, when to escalate to a higher level of care, and what to expect once you make a referral.

Recognizing risk factors, protective factors, and warning signs

These terms often get used interchangeably, but they mean different things and answer different questions.

What are suicide risk factors?

Suicide risk factors answer the question: who is more likely to be at risk, and why? They’re characteristics of a person or their environment that increase the likelihood they will consider, attempt, or die by suicide. No single risk factor predicts what will happen for a given person — they’re weighed together, and they describe heightened risk over time, not what’s happening at any one particular moment.

Risk factors are often divided into two categories: static and dynamic. Static factors are historical facts that don’t change, like a prior attempt or a family history of suicide. Dynamic factors are time-sensitive and clinically modifiable, like a recent loss or an ongoing crisis. Both raise risk, but they call for different kinds of attention: static factors describe a standing baseline, while dynamic factors are what a clinician can actually act on or watch change day to day.

Common static risk factors include:

• Prior suicide attempt(s)
• Mental health conditions — depression, bipolar disorder, personality disorders, substance use disorders
• Family history of suicide
• Chronic or serious physical illness

Common dynamic risk factors include:

• Access to lethal means (firearms in the home, unsecured medication)
• Relationship, academic, legal, or financial losses
• Social isolation, discrimination, or victimization
• An ongoing crisis or acute psychosocial stressor

What are suicide protective factors?

Suicide protective factors answer the question: what helps guard against suicide risk? They’re personal or environmental characteristics that lower someone’s risk of suicide. Protective factors don’t cancel out risk factors on a one-to-one basis, but they change how much support a patient has to draw on when risk is elevated, which is part of why two patients with similar risk profiles can need different responses.

Common protective factors include:

• Strong, stable relationships and social support
• Connectedness to community, school, or work
• Cultural or religious beliefs that discourage suicidal behavior or encourage help-seeking
• Access to effective mental health care
• Problem-solving and coping skills
• A sense of purpose or self-esteem
• Economic stability, including health insurance coverage and access to home internet, which CDC research has linked to lower community-level suicide rates, likely through easier access to care and reduced isolation

What are warning signs of suicide?

Warning signs answer a different question entirely: is someone at immediate risk of suicide right now? Risk factors and protective factors describe someone’s standing likelihood, whereas warning signs indicate immediate risk. The Suicide Prevention Resource Center’s comparison to heart disease is helpful: risk factors for heart disease, like tobacco use and high cholesterol, don’t mean someone is having a heart attack, only that they’re more likely to at some point. Warning signs of a heart attack, like chest pain or shortness of breath, mean it may be happening right now.

The same distinction holds for suicide. A history of a prior attempt is a risk factor; talking about a specific plan is a warning sign. One calls for ongoing attention, and the other calls for an immediate response.

Common warning signs include:

• Talking or writing about death, dying, or suicide
• Expressing hopelessness, feeling like a burden, or having no reason to live
• Increased alcohol or drug use
• Withdrawal from friends, family, or usual activities
• Dramatic mood changes or agitation
• Reckless or risky behavior out of character for the person
• Talking about feeling trapped or in unbearable pain
• Seeking access to lethal means

Being able to differentiate between risk factors, protective factors, and warning signs is crucial. Treating them interchangeably, either escalating every risk factor as though it’s urgent or waiting for a warning sign before responding to accumulating risk, makes both under- and over-response more likely, and it’s what ultimately determines whether a case gets the ongoing attention it needs or the immediate response it requires.

What suicide risk looks like across presentations

Risk doesn't show up the same way in every patient. What's actually driving it changes what a clinician should be watching for. Depression, trauma and PTSD, psychosis, substance use and co-occurring disorders, and personality disorders are the conditions most consistently identified in the clinical literature as carrying elevated suicide risk.

Depression

Depression is the condition most consistently associated with suicide, and it's frequently undiagnosed or undertreated — so risk can be building in a patient who has never been formally diagnosed. Watch for anhedonia (a reduced ability to feel pleasure or interest in things a patient used to enjoy) and hopelessness rather than low mood alone; a patient can present as flat or fatigued without ever describing themselves as depressed. Don't wait for a depression diagnosis on the chart before screening for suicide risk.

Trauma and PTSD

Trauma and PTSD are independently associated with elevated suicide risk: people with PTSD have roughly double the suicide risk of the general population, and that risk climbs further with comorbid depression. Watch for reactivation: risk can spike around anniversary dates, new losses, or situations that echo the original trauma. A sudden return of hyperarousal, intrusive symptoms, or avoidance behaviors after a period of stability often signals that a new stressor has reactivated risk that seemed settled.

Psychosis

Psychosis carries significantly elevated suicide risk, and two features of it change how that risk should be read. First, risk is highest early in the course of illness and in the weeks to months immediately following a psychiatric hospitalization — one meta-analysis found the suicide rate in the first three months post-discharge to be roughly 100 times the global rate, and that risk is especially pronounced for patients with a psychosis or schizophrenia diagnosis. Discharge should be treated as a high-risk transition, not a sign that acute risk has passed.

Second, patients experiencing psychosis may be reluctant or unable to communicate distress the way a clinician would otherwise rely on, so risk can be present without the verbal cues that usually prompt a response. That makes collateral information, from family, prior records, or care team members, more important here than in most other presentations. Command hallucinations — auditory hallucinations that instruct a patient to harm themselves — are associated with suicidal ideation and attempts in some studies, though other research has found no significant link, so their presence should factor into a broader risk picture rather than stand alone as a predictor.

Substance use and co-occurring disorders

Acute intoxication produces disinhibition and impaired judgment that can turn passive ideation into action quickly, which means risk in this population is often tied to active use itself, not just to an underlying condition. Two windows matter clinically: entry into treatment, when substance use is often at its peak and co-occurring depression is common, and relapse during treatment, which clinical guidance identifies as a point where suicide risk can increase, particularly in patients with a prior history of suicidal thoughts or attempts. Treating a substance use disorder and any co-occurring mood disorder together, rather than one after the other, is associated with better treatment outcomes and adherence than treating either in isolation.

Personality disorders

Personality disorders — particularly borderline personality disorder — are often marked by chronic suicidal ideation that can persist for months or years, waxing with stress and waning without it. That's a different clinical picture than a mood disorder, where suicidality is usually tied to an active depressive episode and recedes in remission. The key task is distinguishing a patient's long-standing pattern from an acute escalation on top of it: a patient with a history of suicidal ideation or self-harm isn't automatically at higher risk today than they were last month.

But the literature also documents the opposite failure mode — clinicians underestimating self-injurious behavior or treating chronic disclosures as non-lethal "gestures" precisely because they're familiar, which can desensitize risk assessment over time. The more useful question isn't "is ideation present," which it may chronically be, but "has anything changed."

Across all five, the pattern is the same: risk shifts. A discharge, a relapse, an anniversary, a stressor landing on an existing pattern — reading those shifts matters more than matching a patient's presentation to a generic checklist.

Assessing and responding to risk

Static risk describes how much vulnerability a patient carries; it doesn't say whether today is the day that vulnerability becomes urgent. What determines that is what the clinical literature calls a precipitant: an event tied to humiliation, shame, or despair, such as a relationship loss, a financial or health crisis, intoxication, or family turmoil. A precipitant can escalate a patient with significant standing risk into what the personality-disorder literature calls acute-on-chronic risk, or it can escalate someone with little standing risk just as quickly, from a different starting point. What matters is whether the precipitant has changed the picture, not how much risk was already there. Response should follow that same read.

Chronic risk

Chronic risk calls for consistent follow-up and a current safety plan, not an emergency response triggered by a disclosure that isn't new. In conditions marked by chronic suicidality, like borderline personality disorder, suicidal disclosures often function as a way of communicating distress rather than signaling an immediate plan to act. Treating every one of these disclosures as an emergency, especially through repeated hospitalization, doesn't address what's actually driving the distress, and it can reinforce the behavior instead of resolving it. Researchers call the resulting dynamic "coercive bondage": a clinician, worried about being blamed if something happens, does whatever seems necessary to head off any risk, and the relationship becomes organized around managing threat rather than doing the therapeutic work that would actually help.

The task is to calibrate the response to what's actually changed: a new stressor, a shift in access to means, a recent hospitalization, a break in a support system. If nothing has changed, the response is often to continue the existing plan and document what was asked and what was found. If something has changed, that's what shifts the response, not the fact that the ideation itself is present again.

Acute risk

Acute risk is marked by the same combination across virtually every clinical framework: a specific plan, intent to act on it, and access to the means to carry it out. Passive ideation without a plan is a different clinical situation than ideation with a plan and a timeline. Protective factors matter, but they may not counteract risk at this level; a patient with strong social support and a stable job can still be in an acute crisis if intent and access to means are both present.

A common hesitation, among clinicians and the general public alike, is that asking someone directly about suicidal thoughts might plant the idea or push them toward acting on it. It doesn't: multiple studies, including a systematic review in Psychological Medicine and a 2018 meta-analysis of over 13,000 participants, have found that asking directly doesn't increase ideation or the likelihood of an attempt. An indirect question ("you're not thinking of hurting yourself, are you?") tends to produce a reflexive no, while a direct one gets a more honest answer.

When intent and access to means are both present, one part of the response is directly actionable regardless of anything else in the clinical picture: restricting access to lethal means. This is one suicide-prevention intervention with direct evidence behind it, and it doesn't require resolving the rest of the case first — a clinician, family member, or care team can act on it the same day.

Whether the risk is chronic or acute, a safety plan is not the same thing as a safety contract. A "no-harm contract," where a patient agrees, verbally or in writing, not to harm themselves, is common in practice, often because it feels protective. However, SAMHSA's clinical guidance is explicit that such contracts are never sufficient as a stand-alone intervention and offer no meaningful protection against liability. A safety plan is a different tool, built collaboratively with the patient rather than signed by them, and it's the one with actual evidence behind it.

Risk picture	What it looks like	What the response should be
Chronic, static risk	Long-standing pattern of ideation or self-harm, no new precipitant, no plan, no change in access to means	Consistent follow-up, a current safety plan, reassessment at the next meaningful touchpoint — not an emergency response
Acute risk	A new precipitant, a specific plan, intent, and/or access to means	Immediate action — direct conversation, safety planning, and, if means are involved, restricting access right away
A warning sign or precipitant appears	Direct disclosure, a new stressor, a shift from passive to active ideation	Ask directly ("are you having thoughts of suicide?") — direct questions don't increase risk and get more honest answers than indirect ones
Intent and access to means are both present	Plan, timeline, and unrestricted access to lethal means	Restrict access immediately — this is directly actionable regardless of anything else in the picture
A patient has a "no-harm contract" instead of a safety plan	Verbal or written agreement not to self-harm, with no collaborative plan behind it	Replace it with an actual safety plan — contracts aren't sufficient alone and don't reduce liability

When a higher level of care is necessary and what happens next

Reading risk and responding to it well sometimes points to a clear conclusion: a patient needs more than what's currently available to them. That call is more obvious when a patient is in acute crisis. It's harder when the case sits just below that line: chronic risk that's been stable and then isn't, a warning sign that's new but not yet severe, a patient whose current level of care was adequate a month ago and may not be now. The question isn't "is this patient in crisis"; it's "has what I'm providing stopped being enough for what I'm seeing."

Once that call is made, a common source of hesitation in referring someone out of your care is not knowing what happens next: whether the risk you identified will reach the next clinician or disappear into an intake process with no communication back.

At Charlie Health, that visibility is built into the process itself. Our Admissions Team conducts a comprehensive assessment, weighing the same risk factors and warning signs covered throughout this piece against the patient's current presentation, and puts a safety plan in place before treatment begins if high risk is identified. We keep the referring provider informed throughout, whether the outcome is an admission or help finding the right level of care elsewhere.

Risk assessment doesn't stop at intake. Care is structured around multiple points of contact each week, group sessions, individual therapy, and family therapy, rather than a single weekly touchpoint, so a change in presentation has more chances to be caught. When a warning sign surfaces, it's escalated through the Care Team rather than held for the next scheduled appointment; a screen with no mechanism to act on what shows up afterward isn't actually protective.

When treatment is ending, discharge planning is built into our process. Someone who looks ready to step down can still be carrying distress that hasn't surfaced yet, so the same structure of frequent contact used throughout treatment is used to screen for that before discharge, not just confirm that a patient is ready to leave.

In conclusion

The stakes here are real and worth naming plainly rather than softening or overstating: missing a shift from standing risk to an active crisis, or responding to a chronic disclosure as though it isn't serious, can cost a life. So can the opposite error, treating every disclosure as an emergency, which erodes the trust that makes a patient willing to disclose at all. Neither direction is safer by default.

What's actually protective is paying attention to what's changed, asking directly when something has, and treating a patient's care as continuous rather than assuming stability once a decision has been made.

For a quick, scannable resource to share with staff, families, or anyone in your network, we've put together a flyer for National Suicide Prevention Month covering the warning signs of suicide and how to respond.