College student mental health improves, but significant needs remain

 

Newly published data finds improving college student mental health, persistent barriers to care and a growing support role for faculty and staff.

College students are reporting lower rates of depression, anxiety and suicidal thoughts, continuing a multiyear pattern of improvement, according to recent findings from the 2025–2026 Healthy Minds Study.

The national student report—based on responses from 67,477 students at 107 colleges and universities—found that 34% screened positive for moderate or severe depression, down from 37% in the previous report. Moderate or severe anxiety declined from 32% to 30%, while suicidal ideation fell from 11% to 10%.

Positive mental health, measured through relationships, self-esteem, purpose and optimism, increased from 36% to 43%.

“The data are truly encouraging,” said Justin Heinze, co-principal investigator of the Healthy Minds Study and an associate professor of Health Behavior and Health Equity at the University of Michigan School of Public Health. “Not only are we seeing the continuation of downward trend in screens for major mental health concerns, but we’re also seeing general well-being improve. It’s not a guarantee that decreasing depression or anxiety means an increase in positive mental health, so I’m really pleased to see that prevalence going up—but clearly there is still a lot of need.”

Despite the improvements, mental health challenges remain widespread. One-quarter of students reported non-suicidal self-injury during the past year, 28% screened positive for being at risk for an eating disorder and two-thirds said emotional or mental difficulties had affected their academic performance during the previous four weeks.

Mental health care and barriers

Among students who screened positive for depression or anxiety, 61% received counseling, psychiatric medication or both during the past year. Across all students, 36% received therapy and 30% took psychiatric medication.

Time and cost remained major barriers. Twenty-three percent cited insufficient time, 19% cited financial reasons and 12% said they were unsure where to go for care.

The report also found that 7% of students received counseling or support from an artificial intelligence chatbot such as ChatGPT, reflecting a changing landscape of informal and digital help-seeking.

“AI chatbots are becoming part of the mental health support landscape for college students, and that raises important questions about safety, effectiveness, equity, and how these tools fit into the broader network of care that campuses provide,” said Erin Voichoski, statistician and data manager of the Healthy Minds Study. “We need to better understand what students are seeking from these tools and how they may be able to complement—not replace—meaningful connections with mental health professionals and other sources of support.”

Faculty and staff balance student support with their own needs

A companion Healthy Minds Study report surveyed 5,557 faculty and staff members at 16 institutions. Nearly half—48%—had a one-on-one conversation with a student about mental or emotional health during the previous year. However, only about half felt confident recognizing student distress, and 16% said supporting students had taken a toll on their own mental health.

Among faculty and staff, 16% screened positive for depression, 16% for anxiety and 28% reported high or very high work-related burnout. Only 30% agreed that their institution prioritized employee mental health, compared with 61% who said student mental health was an institutional priority.

Together, the reports show that campus mental health extends beyond counseling centers. It affects academic performance, employee well-being and the faculty and staff members that students often approach for support.

Read the national data reports: Student survey and faculty/staff survey.

School nurses find stock epinephrine improves safety of students with severe allergies

 

Study shows most schools still require families to supply their own

 For school kids with severe food allergies, having epinephrine on hand at school can mean the difference between a scary moment and a life-threatening one. A new national survey of school nurses presented at the 2026 Annual Scientific Meeting of the American College of Allergy, Asthma and Immunology (ACAAI) looks at how schools stock and manage this life-saving medicine. The results showed that while keeping epinephrine on hand at school clearly helps, many families are still being asked to supply their own, even when it may not be necessary.

"Our findings show that having stock epinephrine available at school clearly improves student safety, since schools without it were far more likely to end up without a working device in an emergency," said Mahi Singh, MD, lead study author and ACAAI member. "At the same time, schools also often require students with a known food allergy to bring their own personal epinephrine device each year and leave it with the school nurse. That can be expensive for families and limit accessibility outside school hours compared to backpack storage."

Epinephrine is the only medicine that can stop anaphylaxis, a severe allergic reaction, from becoming deadly. Many schools now keep a supply of epinephrine available, sometimes called "stock" epinephrine, that any student can use in an emergency. Only 10 states require schools to stock epinephrine, but many schools in other states do so voluntarily. At the same time, schools also often require students with a known food allergy to bring their own personal epinephrine device each year. That can be expensive and hard to keep track of for families.

Researchers surveyed nearly 1,850 school nurses across the country between August and December 2025. Of those, 349 nurses, each representing one school, completed the survey.

• 87% of schools kept stock epinephrine on site, but 74% still required students to bring their own personal device as well.
• Most nurses (74%) preferred a "dual coverage" approach, keeping both stock epinephrine and personal devices on hand, while about a quarter (27%) said stock epinephrine alone would be enough. Dr. Singh notes this key point requires further study.
• 72% of schools reported that no student needed epinephrine at all in the past year, and 97% had never needed to use more than two doses in a single day. This suggests many personal devices go unused and may expire.
• Importantly, 10% of nurses said that epinephrine was not available when a student needed it during a real allergic emergency. Schools that did not keep stock epinephrine were more than four times more likely to report this problem.

Families of students with food allergies often pay for a costly new epinephrine device every year to bring to school, even if it's never used and eventually expires unopened. That adds up to expense and hassle for families, on top of the stress of managing a child's allergy. The researchers say their findings support wider adoption of stock epinephrine programs in schools, which could reduce the financial and administrative burden on families while helping ensure epinephrine is always available when a student needs it most.

Abstract Title: Epinephrine Utilization and Management in Schools: A National Survey of School Nurses

Youth and caregiver reports of mental health issues differ

 

Both perspectives needed

Youth and caregivers frequently disagree about psychiatric diagnoses, and relying on only one informant may substantially influence which mental health problems are identified

Psychiatric assessments of children and adolescents commonly incorporate reports from both youth and their caregivers. However, they may identify substantially different mental health problems, potentially influencing diagnosis, treatment planning, and estimates of psychiatric disorder prevalence. 

A new study by researchers at the University of California San Diego School of Medicine and their colleagues examined the extent to which youth and caregiver reports differed across a broad range of psychiatric diagnoses from late childhood through middle adolescence. Previous studies have focused on a limited number of disorders, like depression or anxiety, at single points in time. The study was published on November 11, 2026 in JAMA Psychiatry.

The study measured how often youth and caregivers agree or disagree on psychiatric diagnoses and clinical problems, and how that agreement varies by diagnosis and developmental stage.

The research analyzed reports from 11,868 youth and their caregivers from age 9 to age 17 as part of the UC San Diego-coordinated Adolescent Brain Cognitive Development (ABCD) study, the largest long-term study of children’s brain and behavioral development in the country. 

Youth and their caregivers were asked the same diagnostic questions during separate clinical interviews. The researchers derived a mental health diagnosis based on the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) from each youth’s own report and each caregiver’s report. They then compared those reports and tracked how much they overlapped for each diagnosis across different developmental stages.

The analysis found:

• Disagreement was the norm: When a diagnosis was reported, it came from only one informant, not both, in approximately 65% to 98% of cases, depending on the psychiatric diagnoses and developmental stage.
• Youth were more likely to report major depressive disorder, sleep problems, substance use, suicidal thoughts and attempts and nonsuicidal self-injury. Teens reported depression more often than caregivers did as they got older.
• Caregivers were more likely to report conduct disorders, eating disorders and generalized and social anxiety disorders. These gaps were widest at younger ages and narrowed as teens got older.
• Agreement generally improved across development stages, but remained limited, suggesting that discrepancies persist throughout adolescence.

The findings highlight the importance of obtaining information directly from adolescents rather than relying exclusively on caregiver reports for adolescent psychiatric assessment, according to corresponding author Laika D. Aguinaldo, PhD, MSW, assistant professor of psychiatry at UC San Diego School of Medicine.

“Teens were more likely to disclose private, internal experiences, but parents were more likely to report more externally visible mental health problems,” she said.

The study has implications for understanding the prevalence of psychiatric disorders in young people and for clinical screening practices.

“If we're really thinking about suicidal thoughts, self-harm, substance use and sleep, most parents don't actually know that their kids are experiencing these mental health problems, which is why we need to ask both the teen and the parent,” Aguinaldo added. “If you only ask one, you might miss the opportunity to identify a young person who needs help and to intervene early.”

Future work will investigate factors associated with youth-caregiver discrepancies and whether these differences influence mental health service access and clinical outcomes.

Read the full study: “Youth-caregiver discordance in psychiatric diagnosis and clinical problems in the ABCD study”

As AI moves into the classroom don’t let it bypass teachers’ professional judgement

As universities race to adopt AI tools, a new paper in Frontiers of Digital Education warns that the technology is quietly reshaping the conditions under which teachers make professional judgements—and that without deliberate pedagogical design, it risks making teaching “technically active but normatively thin.”

The conceptual paper, by Ying Ji and Neil Mercer of the University of Cambridge, introduces the concept of the judgement space: the pedagogical condition in which professional judgement is not only exercised but articulated, examined, and sustained through interaction. In AI-supported higher education, they argue, this space becomes a three-party condition, as AI-generated evaluations enter the conversation alongside teacher and student.
The question is not whether AI can evaluate student work—it increasingly can, the authors write. “The question is whether teaching is designed so that AI’s evaluative outputs become material for professional deliberation rather than a substitute for it.”
Drawing on educational philosopher Gert Biesta’s account of education’s three purposes—qualification (knowledge and skills), socialisation (becoming part of disciplinary communities), and subjectification (developing as an autonomous person)—the authors identify a persistent epistemic asymmetry. AI readily captures qualification-domain evidence such as correct answers, error patterns, and performance metrics, whose quality criteria can be defined in advance. Evidence of socialisation and subjectification—whether a student’s hesitation signals genuine struggle, whether a challenge to a peer reflects real engagement—depends on normative, contextual judgements that only a teacher who knows the student, the discipline, and the purposes can make.
This asymmetry, they argue, is not a technological limitation to be solved by better AI, but reflects the nature of those domains, where what counts as quality is itself a normative question. The result: An evidence base systematically weighted toward what is easiest to measure, narrowing the judgement space even without removing the teacher’s authority.
The authors cite evidence consistent with this concern, including a randomised experiment in which learners using generative AI improved essay scores without corresponding gains in knowledge acquisition or transfer—a pattern the study’s authors termed “metacognitive laziness.”
To counteract this, Ji and Mercer advance the dialogic inquiry triangle, a framework developed in school-based practitioner research, extended here to AI-supported higher education. It structures interaction across three dimensions: inquiry INTO dialogue (building analytical capacity to interrogate AI outputs), THROUGH dialogue (collaborative deliberation through “exploratory talk”), and FOR dialogue (orienting toward independent judgement). From this, the authors derive six structural principles for teacher–student–AI interaction—for example, that AI provides data for professional judgement rather than definitive evaluation, and that teaching should preserve occasions for productive struggle rather than automatically resolving difficulty—along with a Professional Judgement Audit for faculty development.
The authors are explicit that the framework is conceptual. The judgement space construct, the structural principles, and the audit are theoretically derived and remain untested in higher education; the triangle has so far been examined only in school contexts. They set out a three-stage research agenda to empirically investigate the judgement space, test which principles matter most across disciplines, and validate the audit as a reflective tool. They also acknowledge that dialogic practices carry cultural and linguistic specificity, and that highly standardised institutions may offer teachers little discretion to create such conditions.
Faculty development in AI-supported environments must extend beyond equipping individual teachers with AI competencies, the authors conclude. “It must address the pedagogical conditions under which professional judgement can function—conditions that can either be enriched or impoverished by how AI is positioned.”

 

---

Journal

Frontiers of Digital Education

DOI

10.1007/s44366-026-0107-x 

Peer Influence in College Applications

 This paper studies how exposure to peers affects college application behavior using neighbors who attend different high schools to mitigate concerns about common shocks. Using granular classroom-level data, the authors find that exposure to peers who have applied to a program increases application rates. Consistent with an information mechanism, there is no additional effect of peers who are admitted. In counterfactuals, the effect is large enough to close urban-rural gaps and gender gaps in applications to selective college programs. 

Study suggests there is no ‘right’ way to teach statistics, giving flexibility to educators

A new study has investigated whether challenges faced by educators in teaching statistics could be reduced by changing how it is taught and revealed that there is no ‘right way,’ giving flexibility to educators to choose how best to teach the subject.

Students studying subjects ranging from psychology and medicine to law are often required to learn statistics, but many find the subject challenging. Two common difficulties are understanding statistical concepts and experiencing anxiety about statistics.

Experts from the School of Psychology at the University of Nottingham have published a new study in Psychology Teaching Review that explored these challenges and whether changing the order in which people encounter equations and worked examples, and the type of example used, affected their learning. The findings suggest that the order and example type did not affect either statistics understanding or anxiety.  This means educators may have considerable flexibility in deciding how to introduce statistical concepts such as standard deviation.

Dr Kieran Lyon from the School of Psychology is lead author on the paper, he said: “We performed this study to find out whether changing the order of teaching materials and the examples used would reduce statistics anxiety and improve understanding.  Neither order or example type affected these outcomes.  This is good news, because it means that either order and example type is equally effective, and educations can choose without worrying that it will negatively affect students' experience or grades.”

Statistics is often introduced through lectures that present abstract equations before students practice applying them to example data. The researchers investigated whether putting the worked example first, or changing the example type, could make statistics easier to understand and less anxiety-provoking.

The researchers conducted an online experiment focused on standard deviation, a measure of how spread-out numbers are around their average. Participants learned the equation for standard deviation and completed a worked example involving calculating the standard deviation of a set of numbers using Microsoft Excel.

The study tested two different teaching orders: equation-first, where participants learned the equation before completing the worked example; and example-first, where participants completed the worked example before learning the equation.

The researchers also compared two types of worked example. One used real-world data on life expectancy from the United Nations, while the other used a deliberately fun and fictional scenario involving the heights of aliens from another planet.

The researchers found that neither the order in which the material was presented nor the type of example used affected participants’ understanding of statistics or their statistics anxiety.

Nikki Dean Marshall, Assistant Professor in the School of Psychology is co-author on the study and added: “Many students approach statistics with anxiety, so educators are always looking for ways to make the subject more accessible. Our findings suggest there is no single 'right' way to teach statistics. Whether teaching begins or ends with equations or whether examples are real-world or simulated, people appear to learn just as effectively which is great news as it gives educators the flexibility to adapt their teaching to the needs of their learners.”

---

Journal

Psychology Teaching Review

DOI

10.53841/bpsptr.2026.32.2.21 

Autistic children benefitted from online literacy instruction during COVID pandemic

 Despite many obstacles presented by students being at home during COVID-19 lockdowns and restrictions, the autistic children involved in this research project still showed significant progress through online instruction.

Flinders University’s Professor Joanne Arciuli, who led the feasibility study, says that developing useful tools for this area of literacy teaching plugs a gap in autistic care.

“Children with autism present with highly diverse and non-uniform reading profiles, with some experiencing difficulties with reading comprehension more than reading accuracy skills,” says Professor Arciuli.

She points to 2021 research of 26 studies involving more than 1200 autistic children, which showed that word reading scores were higher on average than reading comprehension scores.

While the feasibility study was originally designed to test the effects of small-group face-to-face literacy instruction on the reading skills of autistic children in an area of relative sociodemographic inequity, these plans changed due to stay-at-home orders issued mid-way through the study – from group-based to one-to-one instruction, and from in person to online delivery.

“It ended up being a steep learning curve to conduct a trial in an area of socioeconomic disadvantage in Western Sydney during a global pandemic,” says Professor Arciuli.

“The study continued because we had an ethical duty to continue providing services to research participants, so we moved instruction online via telepractice with the interactive web application ABRACADABRA – and this unexpected change offered a unique opportunity to explore the effects of literacy instruction under the same ‘real world’ conditions impacting clinical settings at the time.

“Autistic child participants in our two instruction groups did make some gains in reading abilities compared to a control group, which is interesting because of the mandated disruptions and complex personal and family issues experienced by our participants.”

The outcomes underline the value of a tool such as ABRACADABRA, a free online interactive web application with 33 activities, designed to support children's development in alphabetics (phonemic awareness and phonics), reading fluency, reading comprehension (vocabulary and comprehension strategies) and writing (spelling).

Developed in 2002 by the Centre for the Study of Learning and Performance (CSLP) at Concordia University in Montreal, Canada, the modular ABRACADABRA program can provide highly individualised literacy instruction.

The outcomes produced a surprise. “We hypothesised initially that children in instruction conditions would achieve gains in their reading accuracy and reading comprehension skills, and that children participating in shared book reading activities would achieve greater gains in reading comprehension than children who did not participate in these. However, children who participated in the intervention conditions made significant gains compared with the control group.”

Professor Arciuli says the application of this research could have far-reaching benefits for telepractice delivery of literacy instruction, especially for children with autism in rural or remote areas who cannot access face-to-face instruction.

The research – “ABRACADABRA literacy instruction delivered by speech-language pathologists to children with autism during the COVID-19 pandemic”, by Annemarie Murphy, Benjamin Bailey, Robert Savage, Rauno Parrila and Joanne Arciuli – has been published in the International Journal of Speech-Language Pathology. DOI:10.1080/17549507.2025.2555990.