Emotional Dysregulation and the Gap in Youth Mental Health

Depression is a state that arrives, lasts a while, and lifts, while a screening questionnaire asks about the past two weeks. A young person whose low periods are short and separated by long stretches of doing well will usually be asked during one of the good stretches, and the negative result will be accurate.

Dysregulation is closer to a trait. The episodes themselves come and go, in the way blood pressure readings fluctuate, but the underlying propensity is steadier: how readily someone floods, and how hard it is to come back. That propensity is present between the low periods rather than only during them, which is what would make it findable by asking once a year. It is more stable than mood rather than fixed, since it responds to treatment and shifts with development.

The paths to an act also differ in speed. The depressive route unfolds over weeks, as hopelessness accumulates and the act follows deliberation. The dysregulated route runs from trigger to act in minutes or hours, and studies of people who survived near-lethal attempts find a substantial share reporting less than an hour between deciding and acting.

Instruments for measuring dysregulation already exist. The most widely used is the Difficulties in Emotion Regulation Scale, or DERS, which has been in use for twenty years. It asks about how a person responds when upset: whether they can identify what they are feeling, whether they can hold back an impulse, whether they believe there is anything they can do to feel better. It is brief, free to use, validated across clinical and community populations, and available in versions a parent can complete about a child.

It would not return the same information as a depression questionnaire, which asks how a young person has felt over the past two weeks. The DERS asks what happens when a feeling arrives, a different question, and one that can be answered accurately during a good stretch.

What it does not yet have is a threshold established against self-harm. Establishing one means taking a population where the measure was administered and following those young people forward to see where on the scale the risk actually rises, so that a clinician would know what a score means rather than only that it is higher than average. That work has not been done for emotional regulation.

What the evidence already shows

What justifies that work is the evidence that already exists, and treatment is where it shows up most directly. Dialectical behavior therapy, developed by Marsha Linehan in the 1990s for chronically suicidal patients and adapted for adolescents by Alec Miller and colleagues in the early 2000s, teaches people to tolerate distress and regulate an emotion as it arrives. It reduces suicide attempts and self-harm. What it does not do is reliably reduce suicidal thinking. With depression treatment it runs the other way: the thinking improves, but no study has shown a reduction in suicide. The thinking and the act do not move together. Reducing one does not reduce the other, which is why treating depression well is not the same as addressing the risk.

The component most closely tied to the act has a name. Impulsivity is not one thing, and the research that separates it finds that acting rashly under emotional load, called negative urgency, is the facet most consistently linked to self-harm and suicidality. Across studies of psychopathology generally it shows associations several times stronger than other forms of impulsivity, including the kind measured in ADHD.

The two kinds of impulsivity come apart in treatment. The impulsivity that defines ADHD is present at baseline, without provocation, and stimulant medication addresses it. Negative urgency is a failure that appears specifically under emotional load, and stimulants do not touch it. The two frequently occur together, and from the outside they look the same, a young person who acts without thinking. In a clinical note both are recorded as impulsivity, and the version that responds to medication absorbs the explanation for the one that does not.

Further evidence comes from what the current system records. Researchers at UCLA and Purdue published an analysis of 72,585 suicides in JAMA Network Open in April 2026. The national mortality database codes each death using diagnostic checkboxes and also stores the narrative accounts written by police officers, coroners, and medical examiners. The researchers applied artificial intelligence language models to those narratives and scored them for underlying mechanisms rather than diagnostic categories.

Indicators of clinically relevant emotional distress appeared in roughly 90 percent of cases. The diagnostic codes on those same records had noted a mental health condition in 44 percent and depression in 28 percent. Investigators at the scene described what they saw in enough detail for a model to identify it years later. The coding system built to capture mental health registered less than half of it.

The study is retrospective, and the narratives were written after the deaths by investigators who knew the outcome, so of course people in that situation were distressed. What it establishes is that the distress was recorded and the system did not register it, rather than that it would have been detectable a year earlier.

Screening and routing also work when a health system commits to them. Henry Ford Health built it into routine practice and reported monthly suicide attempts falling from roughly 11.3 to 0.3 per 100,000 patients, and a 2025 study across more than 475,000 patients found suicide rates about 25 percent lower in systems that adopted the model.

Those programs screen for current suicidal thinking, reported at the moment of contact, which reaches a different group than the one proposed here. A young person who is not in crisis on the day of screening does not meet the criterion. So the approach works, and what has not been tested is the same approach aimed at a characteristic that is present between crises rather than during one.

Most of what exists depends on distress becoming visible. Gatekeeper training, peer support, and awareness campaigns work by helping someone notice a young person who is struggling, and the better programs go further, shifting what students think is normal to say out loud. There is evidence that this works, with peer leaders in one randomized trial substantially more likely to bring a friend to an adult. What none of it reaches is the young person who has nothing to say. Not the one who is reluctant, but the one who is doing well by every measure anyone tracks, and for whom the danger is not a visible decline but what happens on a bad night. These are different problems needing different tools, and a school that addresses one has not covered the other.

Why isn't this already being done?

Emotional dysregulation is not a diagnosis, and it appears in ADHD, where an estimated 25 to 45 percent of children with the diagnosis have significant emotion dysregulation, and also in depression, anxiety, trauma, eating disorders, and in people with no diagnosis at all. Since a diagnostic criterion has to distinguish one condition from another, a feature that turns up everywhere fails that test.

That reasoning is still disputed, and Russell Barkley, among the most cited researchers in the field, has long held that deficient emotional self-regulation is a core feature of ADHD rather than an accompaniment to it.

Neither position affects whether the characteristic is worth measuring, because a screen and a diagnosis are not doing the same job. A screen looks for something that predicts harm and can be treated, and for that purpose the number of conditions a feature appears in does not matter.

Blood pressure was measured for decades before it became a diagnosis, and the diagnostic thresholds came out of the accumulated data rather than preceding it. It is also elevated across many unrelated conditions, and so has almost no diagnostic specificity, which has never been a reason to stop taking it.

The remaining obstacles are structural. Screening recommendations are built around diagnoses, so no diagnosis means no recommendation, no billing code, and no coverage. And the evidence needed spans several fields that publish and fund separately, so no one is responsible for the chain that runs from measuring the characteristic to getting someone into treatment. Depression faced none of this and still took twenty-nine years from entering the diagnostic manual to a screening recommendation.

Building the skills before they are needed

Building the capacity to sit with an unbearable feeling without acting on it takes months of practice, and on the night it matters it is either present or it is not. A system that waits for a crisis reaches a young person at the point when there is no longer time to build it.

The window also runs later than most screening does. The parts of the brain that regulate emotion continue developing into the mid-twenties, well past the age at which pediatric care ends, and the suicide rate climbs steeply across exactly that stretch. The most dangerous years arrive shortly after a young person leaves home, when they are furthest from routine care and from anyone who knows their baseline. Asking from around age ten through the early twenties would cover the years when the capacity is still forming and the risk is rising fastest.

What I am trying to do

My aim is to get emotional regulation measured as a matter of routine. The research that would justify it is slow, the vocabulary that would let families ask for it can start immediately, and neither amounts to much without the people who would have to act on it.

There is a version of this that needs no guideline and could start now. A young person diagnosed with depression, anxiety, ADHD, trauma, or a substance use disorder should also be assessed for emotional regulation, because the diagnosis addresses where the thoughts come from and not what happens when one arrives. If regulation turns out to be a problem, it should be treated alongside the condition rather than instead of it, and dialectical behavior therapy is the best-evidenced treatment for that part. Nothing prevents a clinician from doing this today.

The larger version takes longer. Whether a measure of emotional regulation should sit alongside the depression screening at every annual visit is a question for the bodies that write those recommendations, and they move on evidence rather than on reasoning. That evidence does not yet exist, and producing it is the slow half of this work.

Most of what follows a positive result would be skills instruction rather than intensive treatment. Comprehensive dialectical behavior therapy runs roughly six months and is reserved for severe and chronic presentations, and clinicians who deliver it estimate that a small minority of young people in therapy need that level of care. The rest benefit from the skills themselves, which are taught in groups, online, and in family formats that are considerably easier to reach.

The skills are also not narrowly protective. Emotional regulation in childhood predicts school adjustment, academic achievement, and long-term employment. A young person who learns to ride out that kind of episode is better off in ordinary daily ways, and those benefits arrive whether or not a crisis ever comes.

The awareness piece can begin now. Most parents recognize the behavior immediately when it is described to them, but they have never had a name for it, and without a name there is nothing to raise at an appointment. The absence is compounded when a child already carries a diagnosis, because the reactions get absorbed into it. A family with an ADHD diagnosis in hand attributes the intensity to the ADHD, or concludes the child is simply more sensitive than other children, and neither conclusion prompts anyone to ask whether it is something that can be worked on. It usually can be.

Who has to be involved

Four roles bear on whether a young person is identified and treated, and no one of them can do it alone.

The pediatrician has the standing to raise it, the record that follows a child for years, and the annual visit where a screening question would go. What is missing is any prompt to ask, and any room to do it in, since the well visit is already crowded. What would change things is narrow: any young person already carrying a diagnosis is also assessed for regulation, using an instrument that is free and takes a few minutes. An elevated result would not mean a referral to intensive treatment. It would mean the clinician knows something about that patient they did not know before.

Therapists are the next link, and by several accounts the weakest. A pediatrician who recognizes something generally refers to a therapist, and the therapist decides what treatment follows. Clinicians who run dialectical behavior therapy programs report that many general therapists are unfamiliar with it, or believe they can manage the presentation themselves, so a young person can reach appropriate care and still not receive the treatment matched to what they have. The referral is only as good as the person receiving it. What would change things is a therapist who recognizes dysregulation when it presents and knows when the presentation calls for handing the patient on.

The school counselor is often the first professional a struggling student reaches, and sees the student across six or seven hours a day in a way no clinician does. Counselors vary enormously, though. Some are licensed psychologists who know this material and route accordingly. Others have limited clinical training and would not know to ask about emotional regulation or to name a specific therapy, so a school's referral quality depends substantially on who happens to hold the job. What would change things is a counselor who can recognize dysregulation, teach two or three distress tolerance skills in a single session, which requires no clinical credential, and refer with enough specificity that the clinician receiving the student knows what they are looking at.

For a school, the range of what is possible is wide. At the low end, counselors who understand the construct and can teach a few skills, which costs a professional development day. In the middle, a referral map and someone owning the handoff, so the school knows where its students go and whether they arrive. At the upper end, the skills taught to every student as curriculum, which several groups have adapted for classroom delivery. A school can start anywhere on that range.

Parents are the only constant. They see the child across years rather than in single appointments, they know what ordinary looks like for this particular child in a way no instrument captures, and they are the ones who decide whether anything gets pursued. They are also the only ones here whose stake is not professional. What would change things is a parent who knows the term well enough to raise it at an appointment, and who understands that intensity which does not resolve is a thing with a name rather than a personality trait.

What none of them has is a connection to the others. A teacher notices something in March and it reaches no one else. When a referral does happen, the largest single loss occurs between that referral and a first appointment, because nobody owns the interval and nobody counts how many young people arrive.

None of that requires new research. What it takes is someone treating these connections as their responsibility rather than assuming they will form on their own.

What that looks like in practice

Four things follow, roughly in the order they can happen. Material in pediatric offices, reviewed by a clinician and placed where families wait, which reaches parents at the moment the conversation is relevant. Training for primary care through organizations that already do it, including the REACH Institute, which has taken roughly ten thousand clinicians through child mental health curricula. Presentations to therapists and school counselors, reachable through professional associations and continuing education rather than institution by institution. And a referral map, because any of the above creates a problem if there is nowhere to send a young person.

The clinical content has to be delivered by a clinician to land with professional audiences, which means partnering with a psychologist or physician rather than presenting it myself. I can speak to what this looks like from a family's side.

The map is harder than it sounds. Many therapists list dialectical behavior therapy among their approaches, but far fewer run the full model with a skills group, individual therapy, between-session coaching, and a consultation team, which is the version the trials tested. Cost and availability are national problems, so in many places the map would show gaps rather than options.

Where this came from

My son Collin died by suicide in May 2026 at nineteen, weeks after finishing his freshman year at Washington University in St. Louis. He was diagnosed with ADHD in first grade, later with anxiety, and had twelve years of steady, attentive care. Four days before he died his spring grades came in, they were terrific, and he had an internship starting that he was excited about.

Whether any of this would have changed what happened is unknowable. What can be said is that the characteristic can be measured, and that it responds to treatment.