Skip to article

Research & Evidence

What the 2025 CDC Teen Mental Health Data Actually Says

Understand the 2025 CDC high-school mental health statistics, counseling-access findings, survey limitations, and what schools can responsibly conclude.

By Tagapag Team10 min read

CDC found that 33.3% of students in the 2025 national high-school survey reported persistent sadness or hopelessness—down from about 40% in 2023. That represents improvement in one population indicator, but it does not mean a third of teenagers have a diagnosed disorder, that the problem is solved, or that one school program caused the change. 1 2 4

A newly added 2025 question also revealed substantial self-reported unmet need for counseling or therapy. This article explains what each question measures, why denominators matter, and what schools can learn without overstating the evidence. It is research literacy and general education, not clinical diagnosis, treatment guidance or a safeguarding protocol.

At a glance

  • 2025 figures describe U.S. students in school in grades 9–12, not all teenagers.

  • 33.3% answered yes to a specific past-year sadness/hopelessness question; it is not a depression diagnosis.

  • CDC identified improvement on five indicators from 2023 to 2025; the longer decade tells a more mixed story.

  • Of students reporting persistent sadness/hopelessness with valid counseling data, 29.6% received counseling/therapy and 28.8% reported unmet need.

  • Survey self-reports, missing answers and response rates constrain causal and individual interpretations.

  • School connection and access to appropriate adults merit attention, but YRBS did not evaluate Tagapag.

1. What is the 2025 national YRBS?

CDC's Youth Risk Behavior Survey (YRBS) is a nationally representative, school-based survey of students in U.S. grades 9–12, including public and private schools. The 2025 national analytic sample comprised 10,710 usable questionnaires. The CDC published the major 2025 public summary in September 2026 and related peer-reviewed MMWR analyses in October 2026. 1 2 3

It is a snapshot of a defined school-attending population, not a census of all American adolescents and not a longitudinal follow-up of the same students. Survey year and publication date should be stated separately in headlines. 3

2. What exactly did the sadness-or-hopelessness question ask?

CDC asked whether, over the past 12 months, a student felt so sad or hopeless almost every day for two or more consecutive weeks that they stopped doing some usual activities. 33.3% of respondents with valid answers answered yes in 2025. That is substantially more specific than asking whether someone 'felt sad recently.' 2

It is not equivalent to a medical diagnosis of depression or a count of all students who need treatment. A careful summary says about one in three surveyed U.S. high-school students reported persistent sadness or hopelessness, not 'one in three teens has depression.'

3. Did the 2025 survey show improvement compared with 2023?

Yes, on five of six listed mental-health and suicide-risk indicators, according to CDC's 2026 public summary. Persistent sadness/hopelessness, poor mental health, seriously considering a suicide attempt, making a suicide plan, and attempting suicide were classified as improved from the 2023 to 2025 surveys. Injury in an attempt requiring medical treatment showed no significant change across those two years. 1

The rounded sadness/hopelessness rates were 40% in 2023 and 33% in 2025, a decrease of about seven percentage points—not a seven-percent relative decline. The trend describes a population estimate, not improvement for every individual student. 1 4

4. Why do some reports still call the decade troubling?

The comparison window matters. CDC's published trend chart shows roughly 30% in 2015, 40% in 2023, and 33% in 2025 for persistent sadness/hopelessness. CDC classifies the 2015–2025 trend for that indicator as worsened, even though the most recent two-year movement was positive. 1

It is possible for a shorter-term rebound and a longer-term deterioration to be true simultaneously. Avoid headlines that say the problem is either 'fixed' or worsening without interruption. Ask which years and outcomes are being compared.

5. Is 'poor mental health' the same statistic?

No. Poor mental health is a separate YRBS survey measure. In 2025 CDC displayed a rounded 23% rate for that indicator, while persistent sadness/hopelessness was 33%. The questions have different definitions and should not be summed or described as equivalent diagnoses. 1

The CDC trend series for the poor-mental-health item starts in 2021, rather than 2015. An analysis presenting a ten-year historical trend for it would be comparing unavailable survey measures.

6. What are the 2025 suicide-risk indicators?

CDC's 2025 national chart estimates roughly 14% seriously considered attempting suicide in the previous year, 12% made a plan, and 6% reported an attempt. The corresponding rounded 2023 values were 20%, 16%, and 9%. These are separate survey experiences; they are not counts of suicide deaths, medical diagnoses, or clinical assessments of particular students. 1

Serious personal safety concerns require established qualified-adult and emergency support, not interpretation by a peer mentor or reliance on survey averages. This article intentionally does not attempt individual risk assessment or crisis-management instructions.

7. What was new about the counseling/therapy measure in 2025?

The 2025 YRBS added a first-ever national question asking whether students, in the prior year, (a) received counseling or therapy from a health professional, (b) needed but did not receive it, or (c) did not perceive a need. The question included emotions, concentration, behavior and mental health; it was not limited to depression. 2

Among students with valid answers, 18.6% reported receiving counseling/therapy, 15.4% reported an unmet need, and 66.0% reported no perceived need. These are self-reported response choices—not a clinician's judgment of eligibility, adequacy or severity. 2

8. What do the 29.6%, 28.8% and 41.5% figures mean?

These percentages concern only students who reported persistent sadness/hopelessness AND provided a valid counseling response. Within that subgroup, 29.6% reported receiving counseling/therapy, 28.8% an unmet need, and 41.5% no perceived need. Differences from exactly 100% reflect rounding. 2

Do not say '28.8% of all teenagers could not get counseling.' Among all students with valid answers to the counseling item, the unmet-need estimate was 15.4%. The two figures answer different questions.

9. Why do the denominators differ across figures?

Although there were 10,710 usable questionnaires overall, CDC reported 10,483 valid answers for sadness/hopelessness, 8,200 for counseling/therapy and 8,049 valid answers to both. Some students completed site-specific surveys that did not contain every national item; others had missing answers. 2 3

The missingness is explained in the study and should never be described as thousands of students actively refusing to discuss their mental health. A trustworthy chart labels its denominator and describes the question being measured.

10. What sex differences were reported?

In 2025, persistent sadness/hopelessness was reported by 43.4% of female students and 23.9% of male students in the national analysis. Within the sadness/hopelessness subgroup, counseling receipt was 33.2% vs. 23.3%, and unmet need 31.7% vs. 23.8%, for female and male students respectively. 2

These group percentages do not explain why people feel distress or predict who deserves care. Individuals of any sex may need support, and survey demographic categories should not be used to stereotype individual students.

11. What do racial and ethnic differences show?

Within the persistent-sadness/hopelessness subgroup, CDC estimated counseling receipt of 15.7% for Asian students and 15.4% for Black students; the reported unmet-need figures were 38.4% and 33.9%, respectively. The study did not directly determine the reasons for these differences or measure any individual's attitudes toward seeking help. 2

Some subgroup estimates have wide uncertainty intervals. Local schools should examine access barriers and listen respectfully to students rather than assuming that a national group average explains one student's needs. 2 3

12. Can these figures measure treatment quality?

No. CDC specifically notes that its counseling/therapy item does not identify the type, duration, frequency, quality or adequacy of services received. The 'unmet need' response combines possible barriers such as cost, uncertainty about where to seek help or another reason. 2

As a result, the survey cannot tell us how many students obtained effective treatment, how many were denied a particular service, or whether a named program prevented distress. A survey category is not a treatment outcome.

13. How reliable and representative is YRBS?

National YRBS uses sampling, weighting and nonresponse analysis to estimate experiences of students attending U.S. high schools. CDC reports a 30.6% overall response rate for the 2025 national survey, along with 47.0% school participation and 65.2% student participation. The methods report acknowledges that nonresponse bias might remain even after weighting. 3

These are valuable population statistics—not a perfect census or a measurement of every local school. Also, YRBS is cross-sectional: changes and associations cannot alone establish what caused a student's experience, what improved a particular school's outcomes or whether a digital product worked. 2 3

14. Are local district rates the same as national rates?

No. National results combine a representative sample of students from public and private U.S. high schools; separate state and district YRBS samples generally refer to students in regular public schools within their jurisdictions. Participation and available questions vary across locations. 3

A national estimate should never be presented as the measured rate of a specific school. Local staff should use properly governed school feedback, appropriate survey interpretation and established referral resources instead of asking peer mentors to gather private health histories.

15. What can schools responsibly learn from these data?

A practical response is to examine access to qualified adults, the clarity of student help-seeking pathways and the school climate—not to assume one assembly or app can repair a national indicator. CDC's action guide discusses connectedness, mental-health literacy, supportive school environments and staff referral readiness. 5 7

School leaders might ask whether students know where to turn, whether language or scheduling creates barriers, and whether student activities feel accessible. These are program-review questions, not conclusions already demonstrated by the YRBS, and individual clinical concerns should remain with qualified staff. 5 6

16. What can peer mentoring contribute without overclaiming?

Peer mentors can offer ordinary school navigation, encouragement, shared activities and welcoming conversations. CDC describes opportunities to connect and listen to classmates as supportive school practices. The evidence is not a clinical trial of an individual peer mentor, matching method or school app. 6 7

Tagapag can support school-organized peer mentorship, goals and check-ins with staff oversight. The CDC's 2025 YRBS did not study Tagapag; these data cannot prove Tagapag reduces depression, suicide risk, unmet treatment need or any other clinical outcome. Mentors must not be asked to diagnose, counsel or handle safety issues alone.

17. What should a student or family take away personally?

You are not a population statistic. A national survey can signal why schools should pay attention, but cannot tell someone what their own feelings mean. For personal concerns, a parent, trusted school adult, counselor or qualified health professional can help decide what support is appropriate; urgent safety concerns require adult or emergency support, not another student's independent judgment. 5

Related public reading: Loneliness in High School, Supporting a Friend Who May Be Struggling, and School Connectedness Research.

18. Five questions to ask before sharing a CDC mental-health statistic

First: What is the survey year and publication date? Second: Which exact question and recall period does the number represent? Third: Is the denominator all valid respondents or a subgroup? Fourth: Is the finding a trend, association, or experimental causal effect? Fifth: What remains unmeasured—diagnosis, care quality, barriers, or local results? 1 2 3

For verification, begin with CDC's 2025 public findings and the original 2026 MMWR analysis, rather than an unlabeled image or viral summary. When using 2023 figures, label them historical. 1 2 4

Common questions, answered

What percentage of high-school students reported persistent sadness in 2025?

33.3% answered yes to a specific past-year question about extended sadness/hopelessness disrupting activities. This is not a diagnosed depression rate.

Did youth mental health improve from 2023 to 2025?

CDC classified five national mental-health and suicide-risk indicators as improved between those two survey years, while important concerns remain.

Was the 2023 rate around 40%?

Yes. The 2023 rounded national sadness/hopelessness figure was 40%, compared with roughly 33% in 2025.

Does one in three teens have depression?

The survey does not diagnose depression. It measures responses to a defined sadness/hopelessness question among U.S. high-school students.

How many received counseling or therapy?

18.6% of 2025 students with valid answers to the new counseling question reported counseling/therapy with a health professional in the prior 12 months.

What percentage needed counseling but did not get it?

15.4% overall among valid counseling-question respondents; 28.8% among the specific subgroup reporting persistent sadness/hopelessness and answering both questions.

What does 29.6% mean in the CDC report?

Among students reporting persistent sadness/hopelessness with valid service data, 29.6% reported receiving counseling or therapy.

How many students participated in the national 2025 survey?

The analytic sample included 10,710 usable questionnaires, but the number with valid answers differed by question.

Are all U.S. teenagers represented?

No. The national survey represents school-attending U.S. students in grades 9–12, not all adolescents.

Can YRBS prove social media caused mental-health changes?

No. Cross-sectional observational results alone do not determine individual or national causal mechanisms.

Did CDC study Tagapag?

No. The 2025 YRBS did not evaluate Tagapag, any Tagapag mentoring program or product-specific outcomes.

What should a school do with these findings?

Review staff-supported access, student-friendly referrals, school connectedness and local feedback, without claiming that a single program has proven clinical impact.

Sources & further reading

These references inform the discussion. Research on school connectedness or other mentoring programs is not evidence that Tagapag has produced the same outcomes.

Published
View all resources