A SOCIETAL PLAN FOR MENTAL HEALTH, EARLY INTERVENTION, AND PUBLIC SAFETY

Helping people before crisis while protecting individual rights

Public policy framework

Purpose

This plan proposes a practical system for identifying distress and dangerous behavioral change early, connecting people to help before a crisis becomes a criminal-justice matter, strengthening the role of schools and communities, and creating strict safeguards against government overreach. It is designed around a central principle: a person should never lose liberty merely because someone believes that person may have a mental illness. Intervention should be based on observable need, credible risk, functional impairment, abuse or neglect, and clearly defined legal standards—not labels, unpopular beliefs, personality, politics, religion, disability, or social awkwardness.

This document also addresses a particularly difficult case: a child whose distress may be connected to rejection, neglect, conflict, abuse, or lack of attention at home. In such cases, the parent cannot be the only gatekeeper for help. At the same time, schools should not be given unchecked power to diagnose children or override families. The answer is an independent, reviewable system with multiple doors into care and different levels of response.

Core idea

We should stop waiting for either a crime or a psychiatric emergency. Society needs a lawful middle ground between “do nothing” and “call the police.”

Executive Summary

The United States has built strong systems for reacting after a crisis, but weak systems for recognizing and responding to deterioration before it becomes an emergency. A teacher may notice that a student has become isolated, angry, obsessed with grievance, suicidal, frightened, or unable to function. A coworker may watch an adult unravel. A family member may know that something is seriously wrong. Yet the available choices are often too extreme: ignore it, persuade the person to seek voluntary treatment, call 988, call 911, or wait until the legal threshold for involuntary action is met.

The proposed framework creates a graduated response system. It separates mental-health care from criminal investigation, while allowing the two systems to coordinate when there is a concrete safety concern. It also deliberately prevents any single adult—parent, teacher, administrator, police officer, doctor, or government agency—from having unilateral control over a person’s liberty.

The proposed system has eight parts

1. Universal connection and prevention: every child should be known well enough by at least one responsible adult at school or in the community that a major change in functioning is likely to be noticed.

2. Multiple confidential ways to ask for help: students, relatives, teachers, coaches, clergy, employers, friends, neighbors, and individuals themselves should have a place to report a concern without automatically creating a police record.

3. A graduated triage system: ordinary distress receives support; serious deterioration receives professional assessment; credible threats receive behavioral threat assessment; imminent danger receives emergency response.

4. Independent youth safeguards: when a parent is unwilling, unable, or possibly contributing to the problem, a trained independent team can evaluate the child without simply treating the parent as the sole decision-maker.

5. Behavior-based school safety: schools evaluate observable behavior, communications, threats, severe functional change, victimization, and access to means—not whether a student is “weird,” unpopular, disabled, politically unusual, or diagnosed with a mental illness.

6. A treatment pathway that is easier to enter: same-day assessment, mobile crisis response, short-term stabilization, coordinated specialty care for early psychosis, substance-use treatment, trauma care, and family support.

7. Due-process limits on coercion: compulsory treatment or loss of liberty requires a high legal threshold, independent review, counsel, time limits, appeal rights, and periodic reassessment.

8. Accountability and measurement: measure whether people receive help, whether crises repeat, whether students remain connected to school, whether rights complaints occur, and whether the system shows racial, disability, socioeconomic, or other unjustified disparities.

1. Begin With the Right Problem

Mental illness and violence are not interchangeable. A mental-health diagnosis should never be treated as evidence that someone is dangerous. Youth violence is generally associated with interacting individual, relationship, community, and societal factors rather than one single cause. For that reason, a prevention system should focus on behavior, circumstances, deterioration, threats, victimization, hopelessness, substance misuse, acute psychosis, access to lethal means when paired with credible risk, and sudden loss of functioning.

This distinction matters morally and practically. If the system casts too wide a net, people will hide distress rather than seek help. Teachers will be afraid to speak. Parents will see the process as government intrusion. Students will learn that telling the truth about depression or anxiety causes punishment. A prevention system only works if asking for help is normally safe.

The policy rule

No person should be investigated, disciplined, entered into a threat database, denied an opportunity, or deprived of liberty solely because of a mental-health diagnosis, disability, counseling history, medication use, personality, social isolation, unpopular speech, lawful political or religious beliefs, or a request for help.

2. Build a Culture of Connection Before a Crisis

A great deal of prevention happens before professional mental-health treatment begins. The first protective layer is human connection. CDC identifies school connectedness as protective against poor mental health, substance use, and violence. The objective should therefore be larger than simply hiring counselors. Schools should be organized so that students are less likely to become socially invisible.

A practical school connection model

• Each student is assigned a stable adult adviser, mentor, teacher, coach, counselor, or other trained school adult who checks in periodically and remains a point of contact across the year.

• Schools use short, age-appropriate wellness check-ins at reasonable intervals. These are not psychiatric diagnoses and are not disciplinary tools.

• Staff are trained to recognize meaningful changes: withdrawal, hopelessness, fixation on grievance, abrupt deterioration in attendance or hygiene, repeated statements about death, victimization, escalating aggression, paranoia, severe sleep disruption, or other marked changes from the student’s baseline.

• Schools create low-stigma ways for students to seek help for themselves or friends, including anonymous reporting for safety concerns and confidential requests for counseling.

• Every referral should begin with the least restrictive response that reasonably addresses the concern.

3. When the Home May Be Part of the Problem

A prevention model fails if it assumes that every parent is able and willing to recognize a child’s needs. Some parents are deeply supportive. Others may minimize symptoms, reject counseling, ridicule a child, be absent, be overwhelmed themselves, or be a source of abuse, neglect, instability, humiliation, or chronic conflict. None of those facts should automatically remove parental rights. They do mean that the parent cannot be the only door through which a child may receive help.

The independent youth support pathway

When a school employee or another responsible adult observes substantial deterioration or receives a credible concern, the school should be able to refer the child to an independent multidisciplinary support team. The first objective is not to accuse the parent. It is to determine what the child needs and whether the home, school, peer group, online environment, health condition, substance use, bullying, trauma, or some combination is contributing.

• The child is interviewed privately by an appropriately trained professional when legally permissible.

• The parent is normally informed and included, except where immediate notification could reasonably increase risk to the child or interfere with an abuse/neglect investigation.

• The team gathers information from more than one source before making consequential decisions.

• Disagreement by a parent does not automatically prove neglect, and disagreement by the school does not automatically justify state intervention.

• If ordinary counseling or support can solve the problem, the process ends there. If there are credible indications of abuse, neglect, exploitation, or imminent danger, existing child-protection or emergency procedures apply.

• Families receive a written explanation of the concern, proposed services, privacy rules, and appeal or review options, subject to appropriate safety exceptions.

4. Give Teachers Permission to Notice—But Not Power to Diagnose

Teachers often see children for more waking hours than many other adults, which makes their observations valuable. But teachers should not become amateur psychiatrists or law-enforcement investigators. Their role should be to notice, document concrete observations, listen, and refer.

What a teacher should report

Reports should describe facts: what the student said or did, when behavior changed, whether the student has been victimized or threatened, whether school functioning has sharply declined, and whether there are specific statements or actions suggesting self-harm or harm to others. Reports should avoid labels such as “crazy,” “psychotic,” “dangerous,” or “future shooter.”

Protection from retaliation

A teacher or school employee who makes a good-faith referral through the established process should be protected from retaliation by the school system. Parents retain the right to challenge the substance of a referral and to seek review, but they should not be able to prevent staff from reporting a legitimate safety or welfare concern. Likewise, knowingly malicious or discriminatory reporting should carry consequences.

5. Create Four Levels of Response

Every concern should not trigger the same response. The system should use a four-level model so that ordinary adolescent difficulty is not treated as a public-safety emergency.

6. Separate the Help System From the Discipline System

One of the most important design decisions is that asking for mental-health help should not automatically create a school discipline case, criminal-intelligence file, or permanent threat label. Clinical information should remain within health/privacy rules unless a recognized safety exception applies. Threat assessment records should document the behavior that required assessment and should have retention limits and correction procedures.

This encourages truthful disclosure. A student who says, “I am depressed and sometimes wish I would not wake up,” should normally encounter care—not punishment. A student who says, “I am going to shoot a named person tomorrow and I have obtained a weapon,” requires a different response. The system must be capable of distinguishing the two.

7. Use Behavioral Threat Assessment Correctly

For the smaller number of cases involving a credible concern about targeted violence, schools and communities should use trained multidisciplinary behavioral threat assessment rather than profiling. Federal school-safety resources recognize threat assessment teams as groups that identify, evaluate, and address threats or potential threats, often drawing on administrators, counselors, mental-health professionals, and law enforcement where appropriate.

The team should ask questions such as

• What exactly happened or was communicated?

• Is there a specific target, grievance, plan, preparation, rehearsal, or leakage of intent?

• Has behavior escalated over time?

• What stressors, victimization, losses, or destabilizing events are occurring?

• What protective relationships remain?

• Does the person have access to the means necessary to carry out a specific threat?

• What intervention would reduce risk without unnecessarily excluding, criminalizing, or stigmatizing the person?

The purpose of threat assessment is prevention and management, not prediction with certainty. No algorithm should be allowed to declare that a child is a future violent offender. Human review and evidence are indispensable.

8. Make Treatment Easy to Enter

Identifying need has little value if the next available appointment is months away. Every community should have a defined behavioral-health crisis continuum consistent with the national model of someone to contact, someone to respond, and somewhere safe to go. That means 988 or an equivalent contact point, mobile crisis capability, and appropriate stabilization options instead of routing every case through police or an emergency department.

Priority clinical pathways

• Same-day or next-day assessment for serious deterioration.

• Rapid suicide-risk assessment and safety planning.

• Mobile crisis response for people who cannot safely or reasonably reach a clinic.

• Coordinated specialty care for emerging or first-episode psychosis. NIMH-supported research has found better outcomes with coordinated, team-based early treatment, particularly when care begins soon after symptoms emerge.

• Trauma-informed counseling and services for abused, neglected, bullied, or victimized youth.

• Substance-use treatment integrated with mental-health treatment rather than treated as a separate problem.

• Family education and coaching when the family is capable of being part of the solution.

• A non-family advocate or mentor when the family cannot safely fill that role.

9. Create a Trusted-Adult Guarantee for Children

Every child with an identified serious concern should have at least one named responsible adult who is accountable for follow-up. Ordinarily that will be a parent or guardian. When that relationship is unavailable or unsafe, it may be another legally appropriate adult or professional: counselor, social worker, relative, mentor, case manager, or court-appointed advocate where necessary.

The objective is not to replace parents. It is to ensure that a child cannot completely disappear between agencies because each adult assumed someone else was following up.

10. Establish a Clear Follow-Up Rule

Many systems are good at referral and poor at follow-through. Every Level 2 or Level 3 case should have a documented follow-up interval. The responsible professional should verify whether the person actually reached care, whether the situation improved, and whether new information changes the risk level. Cases should be stepped down and closed when appropriate rather than remaining open indefinitely.

11. Protect Civil Liberties With Hard Limits

A system powerful enough to intervene early can also be abused. Safeguards therefore cannot be an afterthought; they must be part of the architecture.

• No mental-health registry of the general population.

• No predictive-policing score based simply on diagnosis, race, disability, poverty, religion, politics, school performance, or social-media ideology.

• No automatic police referral merely because someone seeks counseling.

• No permanent “threat” label without a retention limit and correction/appeal process.

• No involuntary confinement or forced treatment without the legal standard required by state and federal law, except narrowly defined emergency authority followed by prompt review.

• Independent decision-making when significant liberty interests are at stake; the referring teacher, parent, officer, or clinician should not be the sole final decision-maker.

• Access to counsel and judicial review when involuntary detention or treatment is sought.

• Written findings that identify the behavior and evidence supporting major restrictions.

• Regular audits for false positives, discrimination, improper information sharing, and unequal treatment.

• Strong privacy rules and role-based access to records.

12. Handle Parents and Schools When They Disagree

Parent-school conflict should be expected and designed for rather than treated as system failure. A parent may believe a school is overreacting; a teacher may believe a parent is refusing to see a serious problem. Neither side should automatically win.

The resolution mechanism

• Ordinary support decisions remain with the family and school under existing law.

• For a substantial but non-emergency concern, either side may request an independent second-level review by a district or community team not directly involved in the dispute.

• The reviewer examines documented behavior and the child’s welfare, not personalities or which adult is more forceful.

• If the dispute concerns disability rights, special education, abuse/neglect, medical consent, or custody, the existing legal framework for that issue controls.

• If there is imminent danger, emergency safety law governs first, followed by prompt review.

13. Adult Community System

The same philosophy should apply outside schools. Employers, clergy, relatives, neighbors, and friends often notice deterioration but do not know what to do. Communities should have one easy-to-remember entry point for consultation. A person should be able to say, “I am worried about someone and I do not know whether this is an emergency,” and receive guidance without automatically dispatching law enforcement.

Mobile crisis teams should handle appropriate behavioral-health crises; police should remain available when violence, weapons, criminal conduct, or an unsafe scene requires law-enforcement authority. Co-response models may be appropriate in uncertain high-risk cases.

14. Voluntary Psychiatric Advance Planning

People with recurring serious mental illness should be offered a voluntary advance-planning tool while they are well. The person could identify trusted contacts, preferred hospitals or clinicians, medications that have or have not worked, early warning signs, care preferences, and circumstances under which designated people may be contacted. This shifts part of crisis planning from government control to the individual’s own prior choices.

15. What Schools Should Teach

Mental-health education should be practical rather than diagnostic. Students should learn how to recognize when a friend may need help, how to report threats without spreading rumors, how to respond to bullying and isolation, how to seek help confidentially, how substance use affects judgment and mental health, and why asking for assistance is different from getting someone “in trouble.”

Students should also learn the difference between disagreement and danger. Schools should not teach children to report each other for controversial opinions. Reports should concern safety, severe distress, abuse, threats, or significant changes in functioning.

16. Community and Family Supports

Some mental-health problems cannot be solved inside a clinic because the destabilizing condition is loneliness, family breakdown, unstable housing, abuse, unemployment, lack of meaningful activity, or isolation. Communities should strengthen mentorship, youth activities, faith and civic partnerships where voluntarily chosen, peer support, parenting resources, respite care, employment and training supports, and practical assistance for families under severe stress.

This is not a claim that family problems cause every serious mental illness or act of violence. It is recognition that protective relationships and connectedness matter, and that a prevention system should strengthen them rather than waiting until clinical symptoms become catastrophic.

17. What Artificial Intelligence Should—and Should Not—Do

AI could eventually assist with administrative triage, identify missed follow-up, summarize information for trained professionals, or flag that several independent reports concern the same event. It should not be allowed to secretly score the population for future violence or make final decisions about involuntary treatment, school exclusion, arrest, or deprivation of rights.

Any AI used in this field should be auditable, limited to defined purposes, tested for bias and error, subject to human review, and prohibited from making consequential decisions solely on automated inference. The person affected should have a meaningful way to contest incorrect information.

18. Implementation: A Step-by-Step National Model

1. Adopt a model state framework defining the four response levels, privacy rules, due-process protections, and independent review requirements.

2. Require every K–12 district to maintain a multidisciplinary student-support/threat-assessment capability, with training that clearly separates mental-health support from threat assessment.

3. Create a single local access point linked to 988, mobile crisis, community mental-health providers, substance-use care, and youth services.

4. Guarantee rapid evaluation slots for high-priority referrals and develop early-psychosis programs regionally where local population cannot support a standalone program.

5. Establish the independent youth pathway for cases in which a parent may be unable, unwilling, or unsafe to serve as the sole gatekeeper.

6. Create written protocols for parent-school disagreements and appeals.

7. Fund school-connectedness strategies, mentoring, counselors, and training alongside security measures rather than treating security hardware as the entire prevention strategy.

8. Require follow-up and case closure standards so people are not simply referred and forgotten.

9. Create statewide privacy, audit, record-retention, anti-discrimination, and complaint requirements.

10. Publish anonymized outcome measures annually and revise the program based on evidence.

19. Measures of Success

The program should not be judged by how many students it labels, how many people it detains, or how many referrals it generates. Those measures can reward overreach. Better measures include:

• Time from serious concern to professional assessment.

• Percentage of referred individuals who actually connect with appropriate care.

• Repeat crisis contacts and repeat emergency-department use.

• School attendance, connectedness, and successful return after crisis.

• Suicide attempts and serious self-harm among participants.

• Credible threats successfully resolved without violence.

• Use of arrest or exclusion when a health-centered alternative was appropriate.

• Complaints, appeals, overturned decisions, privacy breaches, and improper referrals.

• Disparities in referrals and coercive interventions.

• Family and participant reports of whether the system helped.

20. The Governing Principle

The central challenge is not how to “police mentally ill people.” It is how to notice when a human being is deteriorating, give that person a realistic path back before catastrophe, and recognize when deterioration has become a genuine safety concern—all without creating a society in which government monitors everyone who is different.

A good system must be capable of doing two things at the same time: intervene earlier and use coercion less casually. Those goals are not opposites. Earlier voluntary help, stronger relationships, accessible crisis care, careful behavioral assessment, and reliable follow-up can reduce the number of situations that ever reach the point where force or confinement is considered.

For children, society should presume that parents matter enormously while acknowledging that some children will need a safe door to help that does not depend entirely on parental recognition or approval. For schools, teachers should be empowered to notice and refer but not empowered to diagnose or punish based on suspicion. For government, intervention should become more responsive while remaining constrained by evidence, due process, privacy, and independent review.

The standard should be simple: help early, judge behavior rather than labels, use the least restrictive effective response, preserve human dignity, and require strong evidence before taking away liberty.

Evidence Base and Selected Sources

Centers for Disease Control and Prevention, “Risk and Protective Factors — Youth Violence Prevention.” https://www.cdc.gov/youth-violence/risk-factors/index.html

Centers for Disease Control and Prevention, “School Connectedness Helps Students Thrive.” https://www.cdc.gov/youth-behavior/school-connectedness/index.html

Centers for Disease Control and Prevention, “Mental Health — Adolescent and School Health.” https://www.cdc.gov/healthy-youth/mental-health/index.html

U.S. Department of Education, “Preventing Targeted Violence and Threat Assessment Resources.” https://www.ed.gov/teaching-and-administration/safe-learning-environments/school-safety-and-security/preventing-targeted-violence-and-threat-assessment-resources

U.S. Department of Education, Student Privacy Policy Office, “What is a threat assessment team?” https://studentprivacy.ed.gov/faq/what-threat-assessment-team

U.S. Secret Service National Threat Assessment Center, school and targeted-violence prevention resources. https://www.secretservice.gov/protection/ntac

Substance Abuse and Mental Health Services Administration, “National Behavioral Health Crisis Care Guidance.” https://www.samhsa.gov/mental-health/national-behavioral-health-crisis-care

National Institute of Mental Health, “Recovery After an Initial Schizophrenia Episode (RAISE).” https://www.nimh.nih.gov/research/research-funded-by-nimh/research-initiatives/recovery-after-an-initial-schizophrenia-episode-raise

National Institute of Mental Health, “Early Psychosis Intervention Network (EPINET).” https://www.nimh.nih.gov/research/research-funded-by-nimh/research-initiatives/early-psychosis-intervention-network-epinet

Important legal note

This is a policy framework, not legal advice. Any enacted version would require state-by-state review of parental consent, minor consent, involuntary commitment, child-protection, education, disability, privacy, firearm, health-record, and due-process law, as well as applicable federal constitutional and statutory requirements.

© 2026 Brent Roberson. All rights reserved.