SAMHSA's National Behavioral Health Crisis Care Guidance is built on three foundational elements. Which service does the guidance give as an example of "Someone to Respond"?
Choose an answer, then check it.
Social Work & Human Services · Trauma and Crisis
Crisis Intervention Basics
This is educational material about how crisis services and crisis practice work. It is not clinical guidance, it is not a protocol for responding to a real crisis, and it is not a substitute for professional training and supervision. In the United States, anyone in crisis — or anyone worried about another person — can reach the 988 Suicide and Crisis Lifeline by call or text, 24 hours a day, seven days a week.
Want it in plain words first? Jump to Eli explains — the same idea, no jargon.On this page 9 sections
In 30 seconds
In this field, a crisis names a state rather than a diagnosis: an event or situation has overwhelmed the coping a person normally relies on, and that state is expected to be time-limited. Crisis intervention Short, focused help offered during that state, aimed at safety, stabilization, and connection to continuing services rather than at resolving the underlying problem. Full entry → is the short, focused response to it. The goals are safety, Stabilization Bringing an acute situation to a point where immediate danger has passed and ordinary supports and services can take over. Full entry →, and connection to continuing support — not resolving whatever caused the crisis. In the United States that work now runs through a system SAMHSA organizes in three parts: someone to contact, someone to respond, and a safe place for help.
Why this matters
Crisis work is one of the first places students meet the gap between what a field says it does and what the evidence shows. You will encounter 988, mobile crisis teams, and co-responder programs in field placement long before you encounter anything resembling therapy, and employers will expect you to know how the pieces connect. The system is also being rebuilt in real time: 988 went live in 2022 and SAMHSA finalized new national crisis guidance in 2025. Just as important, several practices taught confidently in this area — single-session debriefing above all — do not survive contact with the research. Learning to check a model against its evidence is the transferable skill here.
The college version
What a crisis is in professional practice
Start with the word itself, because everyday usage and professional usage part company here. In this field a crisis is a state, not a disorder and not a personality trait. The crisis literature describes it as an ordinary human reaction to severe psychosocial stress, or as a situation in which a person is overwhelmed and the coping they normally rely on stops working. Two features follow from that definition. First, a crisis is situational: it can occur in someone with a psychiatric diagnosis and in someone without one, and naming a crisis is not the same as naming an illness. Sociologists of psychiatry note that the crisis framing is deliberately less medicalizing than a diagnostic one, treating what is happening as temporary and episodic rather than chronic, and sometimes as a turning point rather than a breakdown. Second, a crisis is expected to be time-limited, and this expectation shapes everything about the response. If the state is bounded, the intervention can be bounded too. Note the boundary with the neighboring topics in this unit: trauma-informed care concerns how services are designed so they do not re-injure people, and the stress and trauma topic concerns what a traumatic event does to a person over time. Crisis intervention concerns the acute state and the short response to it. A crisis may follow a traumatic event, but a crisis is not itself trauma, and calling every difficult situation a crisis empties the term of the precision that makes it useful.
Where the idea came from, and the critique that came with it
Crisis intervention did not begin as a service system; it began as a theory about when help works best. In 1944 Eric Lindemann published a paper in the American Journal of Psychiatry on the symptomatology and management of acute grief — grief treated as a reaction with a describable course that a helper could respond to. Two decades later Gerald Caplan built on Lindemann's crisis theory in Principles of Preventive Psychiatry (Tavistock, 1964), a foundational text for community mental health, addressed to workers across disciplines whose concern went beyond treating illness. Caplan's argument was economic as much as clinical: a limited number of professionals could accomplish the most by concentrating on people facing a developmental or accidental hazard involving loss, threat, or challenge, because those are the moments at which either growth or breakdown may follow and short-term help therefore has maximum leverage. That is the reasoning behind every crisis line and mobile team that exists today. What is worth noticing is that the skepticism is as old as the idea. A reviewer writing in 1965 warned that the framework was so convenient for community planning that it risked being accepted too uncritically, and asked whether the effects of crisis intervention would simply pass with the crisis. Sixty years later that is still the central research question, and it is the reason this lesson labels frameworks as frameworks.
Models are teaching scaffolds, not procedures
Crisis textbooks organize the work into staged models. The best known is Roberts and Ottens' seven-stage crisis intervention model, published in Brief Treatment and Crisis Intervention in 2005 and presented by its own authors, in the title, as a road map to goal attainment, problem solving, and crisis resolution. This lesson deliberately does not lay out its stages as a numbered procedure, because a stage list read out of context becomes exactly the thing a crisis model should not be: a script applied to a person. Models like this are genuinely useful for teaching, because they give a novice a way to notice what a skilled worker is attending to and in what order. What they are not is a validated procedure. Targeted searching of the peer-reviewed literature surfaces no systematic review establishing that any particular staged sequence produces better crisis outcomes than another. Treat them accordingly: cite them as frameworks, say so in coursework, and be suspicious of any course, agency, or vendor that presents a staged model as though it were an evidence-based protocol. The distinction matters beyond crisis work. A framework organizes attention; an evidence-based practice makes an empirical claim about outcomes. Collapsing the two is how unsupported practices survive for decades.
What crisis response is trying to do
The goals of crisis response are narrow on purpose: safety first, then stabilization, then connection to continuing support. Notice what is missing. Crisis intervention does not attempt to resolve the housing problem, the family conflict, the substance use, or the psychiatric condition underneath the crisis. It is not therapy, it is not a substitute for ongoing care, and its success is not measured by whether the precipitating problem went away. This is why it can be delivered by trained crisis counselors rather than only by licensed clinicians, why contacts are brief, and why the handoff is treated as part of the work rather than as an afterthought. SAMHSA's own description of the longer-term vision for 988 makes the handoff explicit: the aim is to link people who call, text, or chat to community-based providers who can deliver a fuller range of services, such as mobile crisis teams or stabilization centers, along with resources aimed at preventing future crises. A student who understands only this much already avoids the most common category error in the unit, which is judging a crisis contact by the standards of a course of treatment.
The U.S. crisis system as it now exists
SAMHSA's National Behavioral Health Crisis Care Guidance, finalized in 2025 as a pair of documents covering national guidelines and model service definitions, is built on three foundational elements. Someone to Contact means the 988 Suicide and Crisis Lifeline The three-digit U.S. entry point, active since July 2022, routing calls, texts, and chats to a national network of more than 200 local crisis centers. Full entry → and other behavioral health hotlines. Someone to Respond means mobile crisis teams that deliver rapid on-site help and connect people to care, along with outreach teams working on prevention and Postvention Support and outreach offered after a crisis or a death, aimed at those affected and at preventing further harm in the same community. Full entry →. A Safe Place for Help means an array of stabilization services where someone can receive immediate treatment and support. The contact layer is the most built out. The Lifeline began in 2005 as the National Suicide Prevention Lifeline; the FCC issued the order designating 988 on July 16, 2020, requiring carriers to activate the code by July 16, 2022, and the transition happened that July. It is free, confidential, and available 24 hours a day by call, text, or chat, routed through a national network of more than 200 local, independent crisis centers, with Spanish-language text and chat and Deaf and Hard-of-Hearing videophone service added in 2023 and interpretation available in more than 240 additional languages. It is for thoughts of suicide, mental health and substance use crises, and emotional distress generally, and people may contact it about themselves or because they are worried about someone else. Alongside these three elements sit co-responder models, which pair mental health professionals with police officers on qualifying emergency calls — an arrangement that exists largely because deinstitutionalization left police as the default first responders to mental health emergencies.
De-escalation, least-restrictive response, and where the evidence runs out
De-escalation Communication and behavioral approaches used to reduce agitation and restore a workable conversation before any restrictive measure is considered. Full entry → is the professional posture the whole system rests on: communication and behavioral approaches used to reduce agitation, with restrictive measures such as physical intervention, seclusion, or rapid tranquilization reserved for situations where everything else has failed. It is taught everywhere and recommended in practice guidance. It is also, as of the most recent Cochrane review, unsupported by randomized evidence — the reviewers screened 345 citations through April 2016 and found not one trial that met inclusion criteria for psychosis-related aggression. That is absence of evidence rather than evidence of absence, and the honest classroom statement is that de-escalation is standard practice that has not been tested the way a drug would be. The pattern repeats one level up. Nearly two thousand mobile crisis units operate in the United States; a 2025 systematic review found only nine qualifying studies since 2000, with mixed results, and concluded that the literature offers tentative support but is too sparse for firm conclusions — the previous review, in 1995, had found no evidence of effectiveness at all. Where stronger evidence does exist it is worth naming precisely: a recent quasi-experimental evaluation found that a co-responder program reduced involuntary psychiatric detentions by 16.5%, about 370 fewer over two years, with no detectable effect on calls for service, offenses, or arrests. That finding sits directly on the principle of Least-restrictive response The principle of choosing the least intrusive option that keeps people safe, reserving involuntary measures for last resort. Full entry → — respond in the least intrusive way that keeps people safe, and treat involuntary intervention as a last resort. Which raises the legal point students most want a simple answer to and cannot have: authority for involuntary evaluation in the United States comes from state law and institutional policy, and a 2026 scoping review found that U.S. professionals do not even define voluntary and involuntary consistently. There is no national rule to memorize. There is only your jurisdiction's statute and your agency's policy, learned under supervision.
After the contact
A crisis contact does not end when the person is calm. Practitioners are trained to close the loop: document the contact according to agency policy, make the handoff to whatever service carries the work forward, and treat follow-up as part of the intervention rather than a courtesy. Documentation requirements, consent rules, and information-sharing limits are set by agency policy and law rather than by a national standard, which is exactly why this lesson does not state them — students learn them in placement, from a supervisor, for that setting. The last piece is the one courses skip: the workers. Crisis work is emotionally demanding and structurally under-resourced, and supervision exists partly to make that load supportable. It is not only a personal-resilience question. A 2025 survey of all 206 centers in the 988 network, with a 77% response rate, found that 71% of the leaders answering that question said their center was understaffed and 89% reported difficulty finding resources to hire, and the authors concluded that persistent staffing problems could put both staff well-being and service quality at risk. When you evaluate a crisis system, the conditions of the people answering the calls belong in the evaluation.
Eli explains
The same idea, in plain words
Explain it like I’m 10
Everybody has ways of handling hard days: talking to a friend, sleeping on it, making a plan. A crisis is when something is bigger than all of those at once, so the usual ways stop working for a while. Crisis workers are the people whose job starts exactly there. Their job is not to fix the big problem underneath. Their job is to make sure the person is safe right now, help things settle, and then connect the person to someone who can help with the rest. In the United States you can reach that kind of help by calling or texting 988, any hour of any day.
Picture it like this
Think of roadside assistance. When a car dies on the highway, the person who shows up does three things: gets you out of the dangerous lane, gets the car running again or towed somewhere safe, and hands you to a mechanic. Nobody expects a transmission to be rebuilt on the shoulder of a road at night. Crisis response works the same way, and judging it by whether the underlying problem got solved is like blaming a tow truck driver for not being a mechanic.
Where the picture stops working
The analogy breaks in two places. A car is an object being worked on; a person in crisis is a participant with rights, preferences, and a say in what happens next, which is why crisis workers are trained to do the least intrusive thing that keeps someone safe rather than the fastest thing. And a broken-down car really is broken. A crisis is an ordinary human response to being overwhelmed, not a sign that a person is defective — which is precisely the framing crisis theory was built to avoid.
Worked example
Follow one contact through the system, as a description of how the pieces connect rather than as a procedure anyone should follow. A grandmother, worried about her adult grandson after a job loss, texts 988. Her text reaches a trained counselor at a local crisis center in her state's network — the Someone to Contact layer, and note that she is not the person in distress, which the service explicitly allows. The conversation itself is the intervention. Where more help is needed and the community has built the next layer, the center can request a mobile crisis team: two responders who go to the home rather than requiring a trip to an emergency department. Their aim is a settled situation and a warm handoff, not a diagnosis. If more support is needed, a stabilization setting provides A Safe Place for Help. Afterward the team documents the contact under agency policy, the handoff to ongoing services is made, and the responders review the contact with a supervisor — supervision here being about the workers' own load and practice, not a treatment for the family.
Key takeaway
Crisis intervention is short, bounded work aimed at safety, stabilization, and connection to continuing support, delivered through a system SAMHSA organizes as someone to contact, someone to respond, and a safe place for help. The honest version of the topic labels its models as frameworks and states plainly that single-session debriefing does not prevent PTSD.
Quick check
3 questions here, of 5 in this lesson’s practice set. Answers stay hidden until you check.
A student proposes that after any serious workplace incident, all staff should be required to attend one group debriefing session so that PTSD is prevented. What does the research support?
Choose an answer, then check it.
A mobile crisis team spends two hours with a family. By the end the immediate situation has settled and an appointment with a community provider is arranged, but the family's housing and conflict problems remain. A student calls the visit a failure. What is the strongest professional response?
Choose an answer, then check it.
Study tools & related lessonsYou’ll learn to · Common mistakes · Easily confused · Key vocabulary · Related
You’ll learn to
- Define crisis in its professional sense and distinguish it from a diagnosis, from trauma, and from ongoing treatment.
- Explain the goals and deliberate time limits of crisis response, and why crisis intervention is not therapy.
- Describe SAMHSA's three foundational elements of the U.S. crisis system and name a service that sits under each.
- Evaluate the evidence status of common crisis practices, including single-session psychological debriefing, de-escalation, and mobile crisis units.
- Analyze why staged crisis models are best treated as teaching frameworks and why no national rule for involuntary intervention can be stated.
Common mistakes
Believing that a single debriefing session after a distressing event prevents PTSD, and that everyone involved should be required to attend one.
This is the most important correction in the topic. A Cochrane review of eleven randomized trials of single-session psychological debriefing found no reduction in PTSD risk, no reduction in general psychological morbidity, depression, or anxiety, and one trial reporting a significantly higher rate of PTSD at one year among those who received it (odds ratio 2.88). The reviewers concluded that compulsory debriefing should stop. The 2023 VA/DoD guideline likewise finds insufficient evidence to recommend for or against psychological or pharmacological intervention in the immediate post-trauma period for preventing PTSD, and notes that Critical Incident Stress Debriefing did not reduce PTSD incidence at six months. Popularity is not evidence, and a well-meant intervention can do harm.
Treating crisis intervention as brief therapy, and judging a crisis contact by whether the underlying problem was resolved.
Crisis response is deliberately time-limited and aims at safety, stabilization, and connection to continuing support. Unresolved housing, family, or health problems at the end of a contact are expected — they are what the handoff is for. Crisis intervention is not therapy and is not a substitute for ongoing care, and evaluating it as though it were produces the wrong conclusion about both.
Presenting staged crisis models as validated procedures because they appear in textbooks and agency training.
Staged models such as Roberts and Ottens' seven-stage model are teaching frameworks. They organize what a worker attends to; they do not carry review-level evidence that following the stages improves outcomes. Label them as frameworks, and apply the same test to any model: does a systematic review support this, or is it simply widely taught?
Assuming that crisis practices in wide use — de-escalation, mobile crisis units — must be evidence-based, and that involuntary intervention follows one national rule.
Scale and evidence are different things. The Cochrane review of de-escalation for psychosis-related aggression found no includable randomized trials; a 2025 systematic review of U.S. mobile crisis units found nine qualifying studies with mixed results and called the literature too sparse for firm conclusions. And authority for involuntary evaluation comes from state law and institutional policy, with a 2026 scoping review finding that even the terms voluntary and involuntary are used inconsistently. Learn your jurisdiction's law and your agency's policy under supervision; do not generalize from a textbook.
Easily confused
Crisis intervention vs. Psychotherapy or ongoing treatment
Crisis intervention is short, bounded, and aimed at safety, stabilization, and connection to further help; therapy is an ongoing course of care aimed at the underlying difficulty. The first ends in a handoff, the second in a treatment plan.
Someone to Contact vs. Someone to Respond
In SAMHSA's framework the contact layer is 988 and other hotlines, reachable from anywhere in the country by call, text, or chat; the response layer is mobile crisis teams that go to the person when a conversation is not enough.
De-escalation vs. Involuntary intervention
De-escalation is a communication posture used first and is available to any trained worker; involuntary intervention is a legal action authorized by state statute, reserved for last resort, and governed by rules that differ across jurisdictions.
A crisis vs. A traumatic event and its aftermath
A crisis names a person's present state of overwhelmed coping and is expected to be time-limited; trauma names an experience and the longer-term responses to it, which the trauma-informed care and stress, trauma, and resilience topics cover.
Key vocabulary
- Crisis (professional sense)
- A time-limited state in which an event or situation overwhelms the coping a person ordinarily relies on; it describes a situation, not a diagnosis.
- Crisis intervention
- Short, focused help offered during that state, aimed at safety, stabilization, and connection to continuing services rather than at resolving the underlying problem.
- Stabilization
- Bringing an acute situation to a point where immediate danger has passed and ordinary supports and services can take over.
- 988 Suicide and Crisis Lifeline
- The three-digit U.S. entry point, active since July 2022, routing calls, texts, and chats to a national network of more than 200 local crisis centers.
- Mobile crisis team
- A team dispatched to meet a person where they are, rather than requiring them to travel to a hospital or clinic, and to link them with continuing care.
- Co-responder model
- An arrangement pairing behavioral health professionals with police officers on qualifying emergency calls, intended to reduce reliance on law enforcement alone.
- De-escalation
- Communication and behavioral approaches used to reduce agitation and restore a workable conversation before any restrictive measure is considered.
- Least-restrictive response
- The principle of choosing the least intrusive option that keeps people safe, reserving involuntary measures for last resort.
- Psychological debriefing
- A single structured session held soon after a distressing event, historically used in the belief it prevents later disorder; trials do not support that belief.
- Postvention
- Support and outreach offered after a crisis or a death, aimed at those affected and at preventing further harm in the same community.
Sources & references
- National Behavioral Health Crisis Care Guidance — Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services
- 988 Lifeline Timeline — Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services
- 988 Key Messages — Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services
- About the 988 Suicide & Crisis Lifeline — 988 Suicide & Crisis Lifeline (administered by Vibrant Emotional Health under SAMHSA funding)
- Psychological debriefing for preventing post traumatic stress disorder (PTSD) (Cochrane Database of Systematic Reviews, 2002, Issue 2, CD000560) — Rose S, Bisson J, Churchill R, Wessely S; record and structured abstract via PubMed (PMID 12076399)
- VA/DoD Clinical Practice Guideline for the Management of Posttraumatic Stress Disorder and Acute Stress Disorder (June 2023) — U.S. Department of Veterans Affairs and U.S. Department of Defense, Evidence-Based Practice Work Group
- De-escalation techniques for psychosis-induced aggression or agitation (Cochrane Database of Systematic Reviews, 2017, CD009922) — Du M, Wang X, Yin S, Shu W, Hao R, Zhao S, Rao H, Yeung WL, Jayaram MB, Xia J; record and structured abstract via PubMed (PMID 28368091)
- Mobile crisis effectiveness: a systematic review and associated functions and forms framework (BMC Health Services Research, 2025) — Peters RL, Fine J, Newton H, Dalack GW, Watson A, Lich KH; PubMed Central (PMC12801951)
- Emergency mental health co-responders reduce involuntary psychiatric detentions in the USA (Nature Human Behavior, 2026) — Dee TS, Pyne J; PubMed Central (PMC12846909)
- The Seven-Stage Crisis Intervention Model: A Road Map to Goal Attainment, Problem Solving, and Crisis Resolution (Brief Treatment and Crisis Intervention, 5(4), 329-339) — Roberts AR, Ottens AJ; bibliographic record verified via Crossref (DOI 10.1093/brief-treatment/mhi030)
- Book review: Principles of Preventive Psychiatry by Gerald Caplan (Mental Health (London), 24(4), 1965, p. 174) — Guise-Moores S, National Association for Mental Health; PubMed Central (PMC5085475)
- Symptomatology and management of acute grief (American Journal of Psychiatry, 1944; reprinted 151(6 Suppl), 1994, 155-160) — Lindemann E; record via PubMed (PMID 8192191)
- Using Crisis Theory in Dealing With Severe Mental Illness - A Step Toward Normalization? (Frontiers in Sociology, 7, 805604, 2022) — Baumgardt J, Weinmann S; PubMed Central (PMC9218753)
- Effect of crisis resolution team treatment on crisis experience and crisis coping: a multicenter pre-post study in Norway (BMC Psychiatry, 25(1), 1067, 2025) — Holgersen KH, Hasselberg N, Siqveland J, Ruud T; PubMed Central (PMC12593888)
- "Involuntary" and "Voluntary" in Psychiatric, Behavioral, and Mental Health Services: A Scoping Review of Definitions (Journal of Behavioral Health Services & Research, 53(1), 142-164, 2026) — Smart BD, Kalathil K, McCall WV, Munjal S, Kirkendall H, Patel M, Taliaferro A, Yaeger LH, Iltis AS; PubMed Central (PMC12876088)
- Workforce and Staffing at 988 Suicide & Crisis Lifeline Centers (JAMA Network Open, 2026) — Matthews S, Holliday SB, Slover R, Liu J, Breslau J, Purtle J, Kilambi V, Schuler MS, Ramchand R, Cantor J, McBain RK; PubMed Central (PMC13147195)
EliExplains lessons are original prose written from the open, credible references above. See Copyright & Licensing.
Researched 2026-08-18
Educational content only. It is not medical, legal or professional advice. Found an error? Tell us.