The Teen Mental Health Funnel
Two 2 edu-fellas walk into a Mental Health bar (1 of 8 in a series)
I. Introduction
Sean and Mike are trying to grasp: what does the USA teen mental health funnel look like?
We couldn’t find one good source. So we built one.
We combined four major national databases (NSDUH, NCS-A, YRBS, MEPS) and four major commentaries and meta-analyses (Cronin, de Haan, Waumans, Cuijpers).
This memo presents our resulting Teen Mental Health Funnel.
The Appendix describes each source, what it contributes, and why it matters.
Scholarly friends: point us to a better funnel; or point out errors and omissions, etc. This is our first cut.
II. Mental Health Status (100 Teens)
33 teens have a diagnosable depression or anxiety
25 teens have a “subclinical version” (mild)
42 teens are totally fine
Editorial note: We’re not yet commenting on whether that data may over-estimate prevalence or under-estimate it. We’re trying to capture what the “Field” is reporting.
III. Of the 33 Serious Teens
8 depression only
13 anxiety only
12 both
Many teens suffer from both conditions simultaneously.
IV. Who Gets Treated (33 Clinical Teens)
15 receive sustained treatment
3 have brief contact only – the guidance counselor meets a few times, or pediatrician says “let’s keep an eye on it”
15 receive no treatment
V. Of the 15 Untreated Teens, Why Not Treated?
a. NSDUH reports the untreated group as:
7 don’t want it, say they’re fine, feel a stigma, etc
6 lack access - can’t find a therapist, parent doesn’t navigate logistics, can’t pay and no insurance, etc
2 nobody notices the teen’s clinical issue – neither parents nor school
b. A second cut of this data; Cronin’s analysis suggests the adjusted reality is closer to:
9 teens don’t want treatment, say they’re fine, feel a stigma, etc
4 lack access
2 nobody notices
In other words, improving access helps, but many teens still do not engage with treatment.
VI. Of the 15 Treated Teens, What Exactly Do They Get?
7 receive therapy only
6 receive therapy plus medication
2 receive medication only
VII. Of the 13 Who Get Therapy, What Type?
CBT 7
“Supportive” 3 - listening rather than structured skill-building
DBT 1
Whole Family Therapy 1
Other 1
BAT - The “Phonics of Therapy - it works but field doesn’t like it” - pretty close to zero
VIII. How Many of the 13 Complete Versus Dropout?
7 complete
6 stop after a few sessions, perhaps even after 1 session
Why dropout? Various reasons. The leading reason is therapist “alliance.” Does the teen trust the therapist? Feel understood? Feel comfortable? Like the therapist?
IX. Outcomes
A. Of the 7 who get Therapy Only:
1 fully recovers
1 improves
2 no improvement
3 drop out
B. Of the 6 who get Therapy + Meds:
2 fully recover
1 improves
1 no improvement
2 drop out
C. Of the 2 who get Meds Only:
0 fully recover
1 improves
1 discontinues
D. Overall out of 15 treated:
3 fully recover
3 improve
3 no improvement
6 drop out
X. Bottom Line
Out of 100 teens:
33 have clinical disorders
15 receive treatment
3 fully recover
*
Coming Soon in the CTF WTF blog
The FunnelBlue Tribe (Annie Casey) examines our Funnel
Red Tribe (Abigail Shrier) examines our Funnel
Scholarly Reformers examine our Funnel
Mike and Sean naively proposed a way forward that unites all 3
Sean and Mike - BAT and Super BAT
Value Added Therapist
Teens Choose Their Adult Helper experiment
From the top, with Languishing Teens too
*
APPENDIX: Data Sources and Commentaries
I. NSDUH
Citation:
Substance Abuse and Mental Health Services Administration (2024). Results from the 2023 National Survey on Drug Use and Health: Detailed Tables.
What it does:
Nationally representative survey measuring depression prevalence, treatment rates, and reasons for not receiving care.
Why valuable:
This is the federal government’s primary source for estimating how many teens receive treatment.
II. NCS-A
Citation:
Merikangas, K.R. et al. (2010). Lifetime prevalence of mental disorders in U.S. adolescents. Journal of the American Academy of Child & Adolescent Psychiatry.
What it does:
Structured diagnostic interviews establishing prevalence of anxiety, depression, and comorbidity.
Why valuable:
Gold-standard epidemiological estimate. But it’s 2010.
III. YRBS
Citation:
Centers for Disease Control and Prevention (2023). Youth Risk Behavior Survey Data Summary.
What it does:
Large national survey measuring sadness, hopelessness, and emotional distress.
Why valuable:
Captures the large subclinical population not included in formal psychiatric diagnoses. We’ll circle back to this, because this includes many Languishing Teens.
IV. MEPS
Citation:
Agency for Healthcare Research and Quality. Medical Expenditure Panel Survey.
What it does:
Tracks healthcare utilization, including psychotherapy and psychiatric medication.
Why valuable:
Best national data on how teens are actually treated - who gets therapy vs medication.
V. Cronin
Citation:
Cronin, Christopher J et al. "What Good Are Treatment Effects Without Treatment? Mental Health and the Reluctance to Use Talk Therapy." The Review of Economic Studies, Volume 92, Issue 3, May 2025, Pages 1699-1737.
Why valuable:
Shows that improving access increases treatment, but not as much as expected. If you mostly just say “expand access” you’re going to be disappointed.
VI. de Haan
Citation:
de Haan, A.M. et al. (2013). Treatment dropout in child and adolescent mental health care. Clinical Psychology Review.
What it does:
Meta-analysis of psychotherapy dropout.
Why valuable:
Demonstrates that dropout is common, often affecting 30–40% of teens. When people say “Therapy works!” those studies are NOT counting the teens who drop out.
This is problematic. It’s akin to Khan Academy saying “Our math platform works!” and then in a footnote explaining “It works for the 5 to 10% of kids who are willing to stick with it, it does not work for the typical kids who finds it boring and just clicks around.”
VII. Waumans
Citation:
Waumans, R.C. et al. (2023). Dropout from youth mental health care. European Child & Adolescent Psychiatry.
What it does:
Updated systematic review of dropout.
Why valuable:
Confirms dropout remains a major weakness of the treatment system.
VIII. Cuijpers
Citation:
Cuijpers, P. et al. (2023). Psychological treatment of depression: A systematic overview. World Psychiatry.
What it does:
Meta-analysis of hundreds of psychotherapy trials.
Why valuable:
Shows that therapy produces real but modest improvements, with fewer full recoveries than commonly assumed. Again, it goes to the “true efficacy” claim. Mike and Sean meet many people in the therapy sector who seem to rely on glib “CBT works” claims when the more careful scholarship is far more cautious.
IX. r therapist
Bonus citation. Reddit has dozens of therapists each week describing their treatment efforts. These seem like very nice people struggling to achieve their goals. These stories line up with the published evidence.
This from today, for example:
I feel like I’m getting worse at my job
I’ve been a therapist for 5 years. There has been ups and downs. However, lately I feel like I don’t know what I am doing. Every session feels bad, I get a lot of feedback from clients that we aren’t going in the direction they want, and I also feel that myself. I feel like I am currently helping no one, and the results (I measure) seem to back that up.
X. Other “Funnels”
The Depression Treatment Cascade (Pence et al., 2012 and related primary-care cascade literature)
The National Academies “Youth Depression Cascade” (National Academies of Sciences, 2016 report)