How Do You Stop Suicidal Thoughts? Ketamine Can Clear Them in Hours

If You Might Hurt Yourself Today

If you have a method, a weapon, pills, a location, a timeline, a note, or the feeling that you cannot stop yourself, this is a medical emergency. Do not try to solve it by reading. Do not wait for a clinic callback. Do not stay alone.

Tap to call 988, tap to text 988, tap to chat with 988, go to the nearest ER, or tap to call 911 if you are in immediate danger.

If the Thoughts Are Active but You Do Not Have a Timeline

If you have started thinking about methods, but you do not have a fixed time, place, or immediate intent, you are in the danger zone before an attempt. This is the point where a fast biological circuit breaker matters most. Do not wait for a six-week antidepressant trial to rescue you from a crisis that is moving by the hour.

If your intent increases, if you obtain the means, if you pick a time or location, or if you feel unable to stay safe, stop clinic-shopping and go to the ER.

If the Thoughts Are Passive but Relentless

Passive suicidal ideation is not harmless. “I wish I would not wake up,” “I want to disappear,” and “I do not want to live like this” are warning signals, not personality traits. This is the best time to act because you still have enough distance from the thought to make decisions.

Do not minimize the words because you are afraid of sounding dramatic. Clinics need accurate risk information. Passive ideation is often the outpatient lane. Active intent is usually the emergency lane. Say exactly where you are.

If You Are Trying to Help Someone Else

Do not argue with the suicidal thought. Do not debate whether their life is worth living. Do not give a speech about gratitude, family, faith, exercise, or tomorrow. Stay concrete.

If they have a plan, means, timeline, weapon, pills, or locked-door isolation, treat it as an emergency: tap to call 911 or take them to the ER. If they have passive suicidal thoughts without current intent, call a physician-led IV or IM ketamine clinic and ask for an urgent evaluation. Use the same clinical language every time: suicidal ideation, treatment-resistant depression, no current plan, no immediate intent, urgent evaluation.

Fear 1: “They will lock me up and never let me out.”

Fear of involuntary hospitalization is consistently identified in population studies as the most frequently cited barrier to help-seeking among people with active suicidal ideation, outranking cost, distance, and stigma. The image is a locked ward, loss of autonomy, indefinite confinement. It is not accurate, but it is vivid and terrifying, and it is doing real work against you.

Do not promise something you cannot guarantee. Do not say “I promise they won’t hospitalize you.” You do not know that. What you can say is this:

If they are at the stage where you are calling a ketamine clinic rather than an ER, the hospitalization fear is even less warranted — and you can be specific: “A ketamine clinic is not a hospital. It is an outpatient medical appointment. You go in, you get evaluated, you go home.” Specificity is the antidote to the catastrophic image in their head.

Fear 2: “Nothing works. I’ve tried everything. This won’t either.”

This is not pessimism. This is the documented cognitive signature of major depression. Hopelessness — the inability to imagine that the future can be different from the present — is a core symptom of major depression, not a rational assessment of one’s actual treatment options. The organ doing the evaluation is the organ that is malfunctioning. Do not argue with it.

Arguing sounds like this: “You haven’t tried everything. You haven’t tried this.” They will have a counter. You will counter back. Thirty minutes will pass and nothing will have moved.

What reaches this fear is not an argument. It is a shift in what you are asking for:

You are reducing the ask to its minimum: one physical action, not an emotional commitment. The belief can come later. Right now you need their body in a chair in a clinic or ER.

Fear 3: “I don’t deserve help.”

Shame-based refusal is the quietest and most dangerous kind because it often presents as calm. The person is not combative. They keep saying some version of “I’ll be fine” or “I’m not worth the trouble” or “you should worry about someone who deserves it.” This is not a personal judgment. Thomas Joiner’s interpersonal theory of suicide identifies perceived burdensomeness — the belief that one’s continued existence costs more than it contributes — as one of the two primary psychological drivers of suicidal desire. The feeling of not deserving help is a symptom of the same mechanism generating the suicidal thinking.

Do not counter this with a list of their virtues. Do not give the speech about how much they mean to you. That argument lands in a brain that has already decided it is not worth hearing. It bounces off.

What cuts through shame is a factual reframe that removes their self-judgment from the equation entirely:

Shame is an argument. Medicine is not.

Fear 4: “You don’t understand. No one can understand.”

This is a withdrawal signal. It is the person beginning to close off the connection between themselves and everyone who could help them. It is not rudeness. It is the illness narrowing their world down to a point.

The instinct is to prove that you do understand — to offer evidence of empathy, to meet them emotionally. The problem is that any sentence beginning with “I understand how you feel” is immediately testable and immediately deniable. They will tell you that you don’t, and they will be partly right, and now the conversation is about you instead of about them.

What works is staying present without trying to win:

This response does not claim understanding it cannot have. It claims presence, which it can deliver. Stay physically close. Do not leave them alone. Social connectedness and a sustained sense of belonging are among the most robustly documented protective factors against suicidal behavior in the clinical literature. The connection itself is doing medical work even when no words are landing.

Fear 5: “I just need to sleep. I’ll feel better tomorrow.”

Minimization is often the last defense before someone agrees to help — which means it can look like resolution when it is not. The person seems calmer, more reasonable. They have a plan: sleep. The urgency feels like it has passed.

It has not passed. Minimization is the illness buying time.

Do not accept a delay without a concrete next step attached to it:

You are not arguing. You are narrowing the timeline. “Tomorrow morning” is specific enough to foreclose indefinite delay. “I’m staying here” removes the isolation that makes the night dangerous. If they will not agree to a specific next step, that is information. Treat it as escalation, not resolution.

Fear 6: The Locked Door

This is no longer a conversation about persuasion. A locked door with a suicidal person behind it — not responding, or responding in ways that suggest immediate danger — is a medical emergency.

You do not talk someone through a locked door to safety by finding the right words. The door needs to be opened.

If they are responding through the door and their words suggest passive ideation without immediate plan or means, stay talking. Keep your voice calm and continuous. A person in contact with a caring presence — even through a barrier — remains socially connected, and sustained social connection is a documented brake on suicidal crisis escalation. Disconnection is more dangerous than the door.

Say this:

If they stop responding, if their words suggest a plan, means, or action in progress, or if you hear sounds that suggest self-harm is occurring: call 911 now. Say this when the call connects:

Crisis Intervention Team training prepares law enforcement officers specifically for psychiatric emergencies — and in many cities, dispatchers can flag a call for CIT-trained officers or co-responder teams pairing a clinician with a patrol officer. That distinction can change how they arrive and how they communicate. Make the request out loud so it is on the dispatch record.

What You Must Never Say

These are the responses that feel like they should help and do not:

“Think about how this would affect me / the children / your mother.” This loads the person in crisis with guilt they cannot process and that the illness will weaponize. Perceived burdensomeness — the felt certainty that one is a net cost to the people who love them — is already one of the two core psychological states driving suicidal desire according to Joiner’s interpersonal theory. Reminding someone in that state of the suffering their death would cause does not interrupt the crisis. It can deepen it.

“I promise I won’t call anyone.” Do not make this promise. You cannot keep it if their life requires you to break it. A promise you cannot keep destroys the only bridge you have at the moment it matters most. Say this instead:

“You seemed fine yesterday.” Suicidal crises do not follow emotional logic that observers outside the crisis can track. This statement tells the person in crisis that their experience is not legible, not credible, or exaggerated. It closes the door.

“I need you to promise me you won’t hurt yourself.” No-suicide contracts and verbal safety promises extracted during a crisis episode have no demonstrated efficacy as safety interventions and have been specifically argued against in the peer-reviewed literature because they shift responsibility to the person least equipped to carry it. Do not ask for the promise. Ask for the next concrete physical action.

Anything that starts with “At least.” “At least you have a job.” “At least you have a family.” The “at least” construction tells the person that their pain requires context before it qualifies for treatment. The pain is the emergency. It does not need context. It needs a medical intervention.

When to Call 911 Against Their Wishes

You are allowed to do this. You do not need their permission to save their life.

The threshold is not “they are upset with me” or “they told me not to.” The threshold is danger. The American Foundation for Suicide Prevention identifies the following as warning signs requiring immediate action. Call 911 when:

• They have a method available right now — pills, a weapon, a means — and have communicated intent to use it, or you have reason to believe they intend to act.
• They have stopped responding and you cannot verify their physical safety.
• They have given away significant possessions, said goodbye in a way that felt final, or written or shown you a note.
• They have told you — at any point recently — that they have a plan, a time, a place, or that they are going to do it.
• You cannot reach them and you know they are alone.

If any of these are true and you are asking yourself “but what if I’m wrong” — you are not wrong enough for that calculation to matter. A wellness check that turns out to be unnecessary costs embarrassment. Not calling costs everything. The math is not close.

Framing the call as psychiatric — not violent, not criminal — gives responding officers information that can change how they approach the scene, particularly in departments with Crisis Intervention Team-trained officers or co-responder programs pairing law enforcement with a mental health clinician. Make the framing out loud so it is captured in the dispatch record.

After They Get Help: They May Be Angry. That Is the Right Outcome.

A person who survives a suicidal crisis and is angry at you for calling for help is a person who survived. That is the only outcome that matters right now.

The anger will probably be real. “You betrayed me.” “I never asked for this.” “You had no right.” These are normal responses to having autonomy overridden at a moment of extreme psychological distress. They do not mean you were wrong. They mean the person is alive enough to be angry.

Do not demand gratitude. Do not defend yourself at length. Do not withdraw because the response is painful.

Then be there. The repair comes after the crisis. The crisis comes first.

IV Ketamine Has the Strongest Proof Because It Beat Control Treatments on Suicidal Thoughts

IV racemic ketamine is the lead treatment in this page because it has the best suicide-specific data. It has been tested in randomized trials, placebo-controlled trials, and active-control trials using midazolam. It has shown rapid reductions on suicide-specific measures, not just general depression scales.

In an emergency-department randomized trial, 88% of acutely suicidal patients who received IV ketamine reached remission from suicidal ideation within three hours, compared with 33% who received placebo. That is not a mood-score footnote. That is suicidal ideation remission measured inside an emergency setting.

In a multicenter hospital trial of patients with severe suicidal ideation, 63% of patients who received IV ketamine reached full remission by Day 3, compared with 31.6% who received standard care plus placebo. Both groups were hospitalized. Both groups received psychiatric care. The ketamine group still doubled the remission rate.

In a midazolam-controlled randomized clinical trial, IV ketamine reduced suicidal ideation by 4.96 more points than midazolam at 24 hours. The same trial found that 55% of patients receiving ketamine had at least a 50% drop in suicidal ideation by Day 1, compared with 30% receiving midazolam.

The practical conclusion is simple: IV ketamine is the treatment with the clearest human trial evidence for fast suicidal-thought relief. It is the option families should know about first. It is the option hospitals should have built into suicide-crisis protocols years ago.

IM Ketamine Is the Practical ER Option Because It Avoids the IV Bottleneck

IM ketamine does not have the same level of proof as IV ketamine. The weakness is specific: fewer studies, smaller patient pools, less blinding, and no completed large emergency-department randomized trial proving the same effect size as IV.

But IM ketamine solves a hospital logistics problem that IV ketamine creates. It does not require IV placement, an infusion pump, or the staffing needed to run and monitor a 40-minute IV infusion. A clinician can give it as an injection, monitor blood pressure, pulse oximetry, sedation, and dissociation, and move faster in settings where IV access slows everything down.

In a prospective interventional study of acutely suicidal patients, a single 0.5 mg/kg IM ketamine injection cut suicide scores by 78% within two hours. The same study found that 76% of patients reported complete clearance of suicidal thoughts within two hours. At one month, 89% maintained the reduction in suicidal ideation, and 64% stayed stable after one injection.

A psychiatric emergency-department trial-design paper states the operational problem directly: most ketamine suicide studies used 40-minute IV infusions, which can be impractical in a psychiatric emergency department. That is why IM ketamine matters. It could put a fast-acting anti-suicidal intervention into ERs, crisis units, rural hospitals, and low-resource clinics that will never build a full infusion service.

The IM conclusion is narrower than the IV conclusion: the early numbers are strong enough to justify hospital protocols and large randomized trials now. The evidence is not strong enough to rank IM above IV. IV has the trial record. IM has the deployment advantage.

Spravato Does Not Come Close to IV Ketamine for Suicide-Specific Relief

Spravato is not the treatment this page is fighting for in emergency suicide care. It is intranasal esketamine, not IV racemic ketamine. It is delivered through a REMS-certified clinic protocol, not an emergency-room IV or IM suicide-crisis protocol. Its strongest evidence is rapid depression-symptom reduction, not direct clearance of suicidal ideation.

The FDA label makes the limit explicit: Spravato’s effectiveness in preventing suicide or reducing suicidal ideation or behavior has not been demonstrated.

The trial record matches that warning. In ASPIRE I, Spravato improved depression scores at 24 hours but did not produce a separate statistically proven advantage on the clinical suicidality severity scale. In ASPIRE II, Spravato improved depression symptoms within 24 hours, but suicide-specific scores dropped sharply in both the Spravato group and the placebo group. Hospitalization, observation, medication changes, and crisis containment lowered suicide scores in both groups, leaving no clear suicide-specific drug victory at 24 hours.

The adolescent trial showed the same split. Esketamine beat midazolam on depression scores at 24 hours, while suicidality scores improved across all treatment groups without a separate suicide-scale win.

That is the plain-language verdict: Spravato can reduce depression symptoms quickly. It has not proven what IV ketamine has proven in suicide-specific trials. It should not be sold to a suicidal reader as equivalent to IV ketamine. It is not equivalent.

The ER Knows Ketamine. It Usually Does Not Use It for Suicidal Ideation.

Emergency physicians already use ketamine. They use it for procedural sedation, severe agitation, and pain protocols. ACEP has published emergency-department ketamine guidance for sedation and low-dose analgesia, and ACEP psychiatric-patient guidance lists ketamine as an option for immediate sedation of a severely agitated, violent, or aggressive patient.

That is not the same as using ketamine to treat suicidal ideation.

In most ERs, ketamine is a sedation drug or pain drug. It is not a standing suicide-crisis treatment. A suicidal patient who is calm, coherent, and asking for help will usually enter the psychiatric-risk pathway, not a ketamine-treatment pathway.

Why Most ERs Still Do Not Offer It

The failure has several causes.

The Result: The Patient Gets Watched While the Suicidal Thoughts Continue

A suicidal patient may spend hours in a monitored room with no belt, no phone charger cord, no bag, no medications, no privacy, and no ability to leave. That can stop an immediate attempt. It does not necessarily stop the thought, “I still want to die.”

If the patient is agitated, intoxicated, psychotic, or violent, staff may use sedatives, antipsychotics, restraints, or seclusion. Those interventions can reduce danger to the patient and staff. They do not equal IV or IM ketamine treatment for suicidal ideation.

The missing action is specific: a physician order for monitored IV or IM ketamine after the patient meets suicide-risk, medical-safety, blood-pressure, psychosis, mania, substance-use, and consent criteria.

Ask About Ketamine When the Patient Enters the ER

Ask early. Do not wait until discharge.

That chart note will not force the ER to give ketamine that day. It creates a written record that the family asked for a rapid anti-suicidal treatment and the hospital had no protocol. Repeated records can reach the people who can change the system: emergency-medicine chiefs, psychiatry chairs, pharmacy committees, risk managers, hospital administrators, and quality-improvement boards.

Use the ER first when the person cannot stay safe. Once the person is searched, watched, and physically protected, ask why the hospital is not offering IV or IM ketamine for the suicidal ideation that brought them there.

Which Suicidal Patients Could Qualify for Hospital Ketamine?

The target patient is not every depressed person. The target patient is someone who comes to the ER with suicidal ideation and needs rapid medical intervention after immediate physical safety is secured.

An eligible patient has been medically screened, is physically safe under observation, has no untreated overdose, has no unstable medical emergency, and has blood pressure, pulse, oxygen status, and breathing inside the hospital’s treatment limits.

The patient must also be clinically appropriate for ketamine that day. That means no acute psychosis, no active mania, no severe intoxication blocking consent, no uncontrolled medical instability, and no condition the hospital protocol lists as a ketamine exclusion.

Eligibility should also require a suicide-specific baseline measure before treatment. The hospital should document the patient’s suicidal ideation before ketamine, after ketamine, before discharge, and at follow-up. Without that tracking, the hospital cannot tell whether ketamine reduced suicidal ideation or merely sedated the patient.

Which Suicidal Patients Should Be Excluded or Stabilized First?

A suicide-crisis ketamine protocol must name the patients who should not receive ketamine until another danger is handled first.

These exclusions do not weaken the case for ketamine. They make the protocol usable. A hospital cannot build suicide-crisis ketamine care by improvising around overdose, mania, psychosis, blood-pressure spikes, or consent problems at the bedside.

IV Ketamine or IM Ketamine: The Hospital Must Pick the Route Before the Patient Arrives

The hospital must decide before the crisis starts which ketamine route its protocol allows.

One hospital might approve IV ketamine only. That pathway fits patients who can receive IV placement, infusion equipment, nursing coverage, blood-pressure checks, pulse oximetry, mental-status checks, and post-dose observation.

In the BMJ severe-suicidal-ideation trial, patients received two 40-minute IV ketamine infusions at 0.5 mg/kg, given at baseline and 24 hours. That is an IV infusion protocol, not a quick injection protocol.

Another hospital might approve IM ketamine for selected emergency cases where IV access, infusion pumps, or infusion staffing would delay treatment. That pathway fits suicidal patients who meet the same psychiatric and medical criteria but need an injection-based route.

A larger hospital could build two separate pathways: one for IV ketamine and one for IM ketamine. The patient receives one route under one protocol. They do not receive IV and IM ketamine together.

The Ketamine Order Set Has to Be Written Before the Crisis

The order set must tell the clinician what to order without guesswork.

It should specify route, dose range, maximum dose, infusion or injection timing, baseline vital signs, blood-pressure stop rules, pulse oximetry, sedation checks, dissociation checks, nausea treatment, rescue medication rules, fall precautions, and minimum post-dose observation time.

It should also specify who can order ketamine: an emergency physician, psychiatrist, anesthesiologist, critical-care physician, or another credentialed clinician approved by the hospital.

Without an order set, the suicidal patient gets whatever the hospital already has: search, observation, medical clearance, psychiatric evaluation, and boarding. The ketamine evidence never reaches the bedside because no one has authority to order it.

Monitoring Cannot Be Optional

Ketamine for suicidal ideation requires monitored medical care. The protocol must state who checks blood pressure, who watches breathing, who tracks sedation, who evaluates dissociation, who manages vomiting or panic, and who decides whether the patient is safe to leave the treatment area.

Spravato clinics already prove that monitored psychiatric ketamine care can operate in routine practice. The Spravato REMS program requires at least two hours of post-dose observation for sedation, dissociation, respiratory depression, and vital-sign changes.

IV and IM ketamine for suicidal ideation need hospital-specific rules, not a copy of Spravato REMS. But the basic point is already settled: monitored ketamine workflows exist in American psychiatric care. Hospitals need an ER version for suicidal ideation.

Psychiatry, Emergency Medicine, Pharmacy, and Risk Management All Have to Own a Piece

A ketamine protocol dies if every department assumes another department owns it.

If no department owns the protocol, the suicidal patient gets observation instead of ketamine.

Follow-Up Has to Be Scheduled Before Discharge

A hospital ketamine protocol cannot end when the suicidal thoughts drop. A single dose can reduce suicidal ideation fast. It does not guarantee protection after discharge.

The discharge plan should be written before the patient leaves the ER or inpatient unit. It should include a next-day contact, a scheduled psychiatry appointment, a decision about ketamine continuation, a therapy referral, medication review, lethal-means restriction, family contact instructions, and return-to-ER instructions if suicidal intent returns.

The hospital should not give ketamine for suicidal ideation and then discharge the patient with “follow up outpatient.” The protocol has to connect fast relief to the next 24 hours, the next week, and the next treatment step.

What Families Should Ask the Hospital

Do not ask the ER doctor to improvise ketamine outside hospital policy. Ask whether the hospital has a policy.

If the answer is no, hand the physician printed studies from the evidence section at the bottom of this page.

Then ask for the answer in the chart:

The demand is not bedside experimentation. The demand is a written hospital pathway for a fast anti-suicidal treatment that outpatient ketamine clinics already know how to monitor and that most ERs still do not offer to suicidal patients.

1. If I might hurt myself today, should I call a ketamine clinic or go to the ER?

Go to the ER now or tap to call 911. If you have a weapon, pills, a plan, a timeline, a suicide note, or the feeling that you cannot stop yourself, outpatient ketamine is not the first step. Physical safety comes first.

A ketamine clinic can evaluate passive or non-imminent suicidal ideation. It cannot replace emergency care when you might act today. In immediate danger, use the ER, tap to call 988, tap to text 988, or tap to chat with 988.

2. How fast can IV ketamine reduce suicidal thoughts?

IV ketamine can reduce suicidal ideation within hours in controlled trials. In one emergency-department trial, 88% of acutely suicidal patients receiving IV ketamine reached remission from suicidal ideation within three hours, compared with 33% receiving placebo.

In a larger hospital trial, 63% of severely suicidal patients receiving IV ketamine reached full remission by Day 3, compared with 31.6% receiving standard care plus placebo. These are not six-week antidepressant timelines. These are hour-to-day timelines.

3. Can ketamine help passive suicidal thoughts?

Passive suicidal thoughts still need medical attention. “I wish I would not wake up,” “I want to disappear,” and “I do not want to live like this” are suicidal ideation even when there is no plan or intent.

Passive suicidal ideation is often the clearest outpatient ketamine lane because the person may be safe enough for clinic screening instead of ER admission.

4. Can I get ketamine if I am actively suicidal?

It depends on the level of danger. If suicidal thoughts are active but there is no immediate plan, timeline, weapon, pills, or intent to act today, a physician-led ketamine clinic may evaluate you urgently.

If you have a plan, means, timeline, or cannot stay safe, the ER comes first.

Most hospitals will not. Ask anyway, and ask them to document the answer in the chart.

5. Is IV ketamine better than Spravato for suicidal thoughts?

Yes, for suicide-specific evidence. IV ketamine has stronger proof for rapid reduction or remission of suicidal ideation. Spravato nasal spray can reduce depression symptoms quickly, but its label states that its effectiveness in preventing suicide or reducing suicidal ideation or behavior has not been demonstrated.

Large Spravato trials showed rapid depression improvement, but suicide-specific scores often dropped in both the Spravato and placebo groups after hospitalization and intensive standard care. That is not the same result as IV ketamine trials showing direct separation on suicidal-ideation outcomes.

6. Is one ketamine treatment enough to stop suicidal thoughts for good?

Usually no. A single IV ketamine treatment can reduce suicidal ideation fast, but the effect can fade. The stronger plan is rapid relief followed by a stabilization block, follow-up care, medication review, therapy, lethal-means restriction, and a return-to-ER plan if suicidal intent comes back.

Repeated-treatment data support this. In one repeated-infusion study, anti-suicidal response rates reached 81.7% by Day 13 and remained 73.3% at Day 26. Fast relief matters. A follow-on plan decides whether relief turns into stability.