A Note on the Evidence Base

This case study addresses a question that rarely gets asked directly in school leadership circles: can school itself be a source of harm to young people’s mental health?

It is a politically sensitive question. It risks being read as an attack on teachers, a justification for poor attendance, or a politically motivated critique of the examination system. None of those readings are supported by the evidence, and none are the argument this case study makes.

Meta Pedagogy produces independent, evidence-informed analysis for school and trust leaders. This document follows the same methodology as our other case studies in this series: claims are grounded in peer-reviewed research and official UK data, the strength of the evidence is clearly signalled throughout, and practitioner commentary is explicitly labelled as such so readers can distinguish between what the research shows and what two decades of working across schools and MATs suggests.

The evidence base on this question is substantial and largely consistent. Academic pressure, exclusion, bullying, poor staff-student relationships, and difficult transition points all carry documented associations with poor adolescent mental health. The picture is not one of schools as uniformly harmful institutions. It is one of school environments that can protect or harm depending on culture, leadership, and systemic pressures, many of which originate outside the school gate.

A MAT CEO or senior leader reading this document should find it challenging, useful, and actionable. That is the intention.

The Problem School Leaders Face

Mental health referrals among secondary-age young people have risen sharply since 2017, with NHS and CAMHS data consistently showing more than a doubling of presentations nationally, though the exact multiplier varies by region. Around one in five children aged 8-16 now has a probable mental disorder, with rates among 17-22 year olds reaching approximately 22%, figures that have risen substantially since 2017. School leaders are routinely told that social media, family breakdown, and post-pandemic disruption are the primary drivers.

They are less routinely told that school itself appears in the evidence as a significant contributing factor for a meaningful proportion of those young people.

That is not a comfortable finding. It is, however, a finding that a growing body of UK research supports, and one that school leaders who are serious about student wellbeing cannot afford to ignore.

Academic Pressure and High-Stakes Testing

The evidence that exam pressure harms a subset of students is robust.

The Children’s Society Good Childhood Report (2023) found that children in the UK report lower levels of life satisfaction than peers in most other developed nations, with school-related pressures consistently identified as a primary driver. England’s young people ranked among the least happy at school of any country surveyed by the OECD’s PISA study (2022), with performance anxiety and fear of failure cited repeatedly in qualitative data.

Research consistently documents clear associations between high-stakes testing environments and elevated anxiety in adolescents, particularly around GCSE and A Level periods. A substantial body of this work is summarised in Putwain and Daly (2014, British Journal of Educational Psychology), which established test anxiety as a significant and independent predictor of underperformance and psychological distress in secondary school students. A 2023 survey by YoungMinds found that 76% of young people said that school or college was a factor in their mental health difficulties, with exam pressure the most frequently cited cause.

A longitudinal study by Putwain et al. (2021, University of Cambridge) found that performance-contingent self-worth, the degree to which young people tie their sense of value to academic outcomes, was a significant predictor of anxiety and depression in secondary school students. England’s examination system, which concentrates high-stakes assessment into relatively narrow windows at age 16 and 18, creates structural conditions that amplify this risk for vulnerable students.

The evidence is not uniformly negative.

Academic challenge, when experienced as stretching rather than threatening, is associated with positive outcomes including resilience, self-efficacy, and motivation. The problem is not examination or high standards per se. It is the absence of adequate emotional scaffolding around them, and a system that provides limited alternative pathways for students whose strengths do not align with traditional academic assessment.

School Culture, Belonging and Exclusion

Belonging is one of the most powerful protective factors in adolescent mental health, and schools have significant influence over it.

A systematic review by Allen and Kern (2017), frequently cited in UK educational psychology literature, found that school belonging, the degree to which students feel accepted, respected, and included, was a consistent predictor of mental health outcomes across cultures and age groups. Students with high school belonging showed lower rates of depression, anxiety, and suicidal ideation. Students with low school belonging showed the reverse.  The practical implication is significant, school culture is a mental health variable.

Exclusion carries some of the strongest documented harm associations in the literature.

Permanent and fixed-term exclusions are associated with substantially elevated risk of mental health problems, criminal justice involvement, and long-term poor outcomes. The Timpson Review (2019) found that excluded students were significantly more likely to have unmet SEND needs, experience of trauma, or mental health difficulties that had not been adequately identified or supported before exclusion. In many cases, exclusion is both a consequence of unmet mental health need and a cause of further deterioration.

Exclusion rates in England rose sharply from 2017 and have remained at sustained highs through 2024-25, with DfE data recording 335,700 suspensions in Autumn 2024 alone. Black Caribbean pupils are three times more likely to be permanently excluded than their white peers, a disparity confirmed in DfE data through 2024-25 that has persisted for decades and points to systemic rather than individual factors.

Bullying remains a pervasive and under-addressed harm.

NHS Digital data (2023) found that children aged 11-16 with a probable mental disorder were five times more likely to have been bullied than those without. A meta-analysis by Gini and Pozzoli (2009), replicated in subsequent UK studies, established bullying victimisation as a significant independent risk factor for depression and anxiety in adolescents. More recent UK data from the Anti-Bullying Alliance (2023) suggests that around a third of secondary school students report being bullied in any given year, with rates higher among students with SEND, LGBTQ+ students, and those from minority ethnic backgrounds.

The evidence on school responses to bullying is less encouraging. A systematic review by Gaffney et al. (2021) found that whole-school anti-bullying interventions produce modest but real reductions in bullying when implemented with fidelity, but that implementation quality in UK schools is highly variable.

Staff-Student Relationships and Pastoral Systems

The quality of staff-student relationships is a significant and modifiable predictor of mental health outcomes.

A substantial body of research, summarised in the EEF’s guidance on social and emotional learning (2021), identifies positive teacher-student relationships as one of the highest-leverage factors in both academic and wellbeing outcomes. Students who report feeling known, respected, and supported by at least one adult in school show substantially better mental health outcomes than those who do not.

This finding has direct implications for how pastoral systems are designed. Schools that have moved towards form tutor models with high caseloads and limited pastoral time, or that have centralised pastoral care in specialist staff with limited classroom presence, may be inadvertently reducing the frequency and quality of the adult-student relationships that the evidence identifies as protective.

The mental health workforce in schools is insufficient and unevenly distributed.

NHS England data shows that the majority of secondary schools have access to some form of mental health support, but the depth, consistency, and quality of that support varies enormously. Mental Health Support Teams, introduced through the NHS Long Term Plan, cover a growing but still minority proportion of schools. Waiting times for CAMHS referrals mean that students identified as needing specialist support frequently wait months or years to receive it, during which time the school remains the primary support structure.

A 2023 survey by the National Education Union found that 90% of teachers reported having supported a student with a mental health problem in the past year, but only 35% felt adequately trained to do so. The gap between the demand placed on school staff and their capacity to meet it is itself a source of harm, both for students who receive inadequate support and for staff whose wellbeing is affected by the weight of unmet need.

Transition Points

The transition from primary to secondary school is a documented risk period.

A longitudinal study by Topping (2011), updated in subsequent UK cohort studies, found measurable increases in anxiety and reductions in school belonging in the first year of secondary school, with effects persisting for students who did not establish strong peer relationships by the end of Year 7. The structural features of secondary school, multiple teachers, subject specialisation, larger peer groups, and more complex social hierarchies, create conditions that are genuinely more challenging for many young people than primary school.

Students with SEND, those from disadvantaged backgrounds, and those with pre-existing mental health vulnerabilities show disproportionately negative outcomes at this transition point, suggesting that the universal experience of secondary transition interacts with existing vulnerability in ways that schools can identify and respond to but frequently do not.

The GCSE-to-post-16 transition is a secondary risk point that receives less attention.

Students leaving Year 11 face simultaneous pressure from examination results, decisions about post-16 pathways, and significant social disruption as peer groups fragment. For students whose results do not meet their expectations or the requirements of their intended pathway, this period carries elevated risk of acute mental health difficulties. CAMHS referral data consistently shows a spike in presentations following GCSE results in August, a pattern documented across multiple years without generating proportionate systemic response.

What Protects: The Evidence on School as a Positive Factor

A balanced assessment requires equal attention to the evidence on schools as protective environments, and that evidence is substantial.

Students who are in school, engaged, and connected to at least one trusted adult consistently show better mental health outcomes than those who are not. School refusal and persistent absence are among the strongest predictors of poor long-term mental health and life outcomes in the literature, a finding that cuts against any simple narrative of school as primarily harmful.

The What Works Centre for Wellbeing and the EEF both identify evidence-based whole-school approaches to wellbeing, including universal PSHE provision, staff training in mental health awareness, and structured peer support programmes, as producing measurable positive outcomes. However, it is important to note that not all wellbeing interventions are equally beneficial. Some programmes, including certain group mindfulness approaches, carry a risk of harm for vulnerable subsets of students through peer contagion effects, where shared focus on distressing experiences within a group setting can amplify rather than reduce anxiety. Implementation quality and careful targeting matter as much as the intervention itself.

High-quality teaching is itself a protective factor. Students who experience consistent, respectful, and intellectually stimulating teaching report higher levels of school belonging, lower anxiety, and better self-efficacy. The relationship between teaching quality and student wellbeing is not always foregrounded in the mental health literature, but it is present across multiple large-scale studies.

What This Means for School Leaders

School is neither uniformly harmful nor uniformly protective. It is a context in which both are possible, and leadership decisions about culture, pastoral systems, assessment scaffolding, and staff development are among the most significant determinants of which direction it tips for individual students.

The students most at risk from school-related harm are largely identifiable in advance. Students with SEND, those from disadvantaged backgrounds, LGBTQ+ students, students from certain ethnic minority groups, and those with pre-existing mental health vulnerabilities show disproportionate harm across almost every mechanism identified in this case study. A universal approach that does not account for differential vulnerability will systematically under-serve these groups.

Not all wellbeing interventions are safe for all students. The evidence on iatrogenic risk, where interventions intended to help cause unintended harm, is growing. School leaders should ensure that any programme adopted has an evidence base that includes attention to potential adverse effects, particularly for students with existing mental health vulnerabilities.

Systemic pressures matter as much as individual school decisions. The structure of high-stakes assessment in England, the rate of exclusions, the underfunding of CAMHS, and the insufficient mental health training provided to teachers are not problems that individual school leaders can solve. They are, however, problems that well-informed school leaders can mitigate, advocate against, and refuse to amplify through their own institutional choices.

Recommended Actions

1. Audit school culture for belonging indicators Use validated tools, the Wellbeing Measurement Framework (NHS England) or the Pupil Attitudes to Self and School survey are both established options, to establish a baseline measure of school belonging across year groups. Disaggregate data by SEND status, ethnicity, and free school meal eligibility to identify where belonging deficits are concentrated. Owner: Deputy Head Pastoral | Term 2, 2025/26 | Low cost | Anchor: DfE Wellbeing for Education Programme
2. Review exclusion practice against SEND and vulnerability data Audit the last three years of fixed-term and permanent exclusions against SEND, mental health, and disadvantage data. If exclusions are disproportionately concentrated in identifiable groups, treat this as a systemic signal requiring curriculum, pastoral, and training responses rather than individual case management. Owner: SENCO/DSL | Term 2, 2025/26 | Staff time | Anchor: Timpson Review (2019); KCSIE 2025
3. Strengthen transition support at Year 7 and post-16 Introduce or strengthen structured transition programmes that extend beyond the first half-term of Year 7, with particular attention to students with SEND and those from disadvantaged backgrounds. Implement a parallel programme for Year 11 students at risk of poor outcomes at the post-16 transition. Owner: Head of Year 7/Head of Sixth Form | From September 2026 | Low to moderate cost | Anchor: EEF Transition Guidance
4. Invest in staff mental health literacy Ensure all teaching and support staff complete recognised mental health awareness training, Mental Health First Aid or equivalent, with refreshers every two years. This is not a substitute for specialist provision; it is the infrastructure that makes early identification and appropriate referral possible. Owner: CPD Lead | Academic year 2026/27 | Moderate cost | Anchor: NHS Long Term Plan; DfE Mental Health in Schools guidance
5. Build emotional scaffolding around examination periods Develop a structured programme of pastoral and curriculum support around GCSE and A Level examination periods, including explicit teaching of anxiety management, realistic appraisal of outcomes, and clear post-results support pathways. This is distinct from reducing academic rigour; it is providing the emotional infrastructure that rigour requires. Owner: Head of Pastoral/PSHE Lead | From Easter 2026 | Low cost | Anchor: YoungMinds School Pressure guidance; PSHE Association (2023)
6. Scrutinise wellbeing interventions for evidence of harm as well as benefit Before adopting any new wellbeing programme, require evidence that adverse effects have been considered. Group-based interventions in particular should be reviewed for peer contagion risk among vulnerable students. Where uncertainty exists, pilot with careful monitoring before trust-wide rollout. Owner: Deputy Head Pastoral/Research Lead | Ongoing | Staff time | Anchor: EEF Social and Emotional Learning guidance (2021)

Meta Pedagogy Commentary

The following reflects practitioner judgement based on 20-plus years working across schools and MATs. It is clearly labelled as such.

The most important thing I want school leaders to take from this case study is not that school is harmful. Most schools, most of the time, are doing an enormous amount of good for the young people in them. What I want them to take is that the question of whether their school is harmful to specific groups of students is one they should be asking routinely and systematically, and that most are not.

The exclusion data is the area that concerns me most in current practice. I have worked with MATs where exclusion rates have risen year on year, where the data shows clear disproportionality by ethnicity and SEND status, and where the response from leadership has been to treat each exclusion as an individual case rather than as a systemic pattern requiring systemic response. The Timpson Review set out very clearly what the evidence says about exclusion and unmet need. It was published in 2019. The practice in many schools has not materially changed.

The examination pressure finding is one that many school leaders privately agree with and publicly resist engaging with, because the structural response, meaningful reform of high-stakes assessment at 16 and 18, is beyond any individual school’s control. What is within their control is the pastoral and curriculum scaffolding around those examinations, and in my experience that scaffolding is thinner than it should be in most schools, particularly for students whose relationship with academic assessment is already fragile.

The iatrogenic risk point is one I want to flag specifically because it runs counter to the instinct of well-meaning leaders. The impulse to do something about student mental health is entirely right. But some of the things schools are doing, rolling out whole-year-group mindfulness programmes, running assemblies that centre on anxiety and stress, commissioning external providers with weak evidence bases, carry a genuine risk of harm for the students most in need of support. Intervention quality and targeting matter. Doing something is not always better than doing nothing.

The staff capacity finding is the one that makes me most concerned about the next five years. The gap between the mental health support that students need and the capacity of school staff to provide it is widening, not narrowing. Teachers are being asked to carry clinical weight they were not trained for and should not have to carry alone. The schools that are managing this well are those that have invested in genuine pastoral infrastructure, not just a designated mental health lead with a caseload of 400 students, but a whole-staff culture in which every adult in the building understands their role in student wellbeing and has the training and support to fulfil it.

The transition finding is one that I think is systematically underinvested in across the sector. Year 7 transition programmes that end in October of the first year are not transition programmes. They are welcome events. The evidence on what works is clear: sustained, differentiated, and extending well into Year 8 for vulnerable students. Very few schools do this.

None of this requires new money in most cases. It requires leaders who have read the evidence, taken it seriously, and made different decisions about where their existing resources go.

Evidence Validation Appendix

For governance and audit purposes. Each key claim mapped to its supporting source and confidence rating.

Claim	Source	Confidence
Around one in five children aged 8-16 has a probable mental disorder	NHS Digital (2023)	High
Rates among 17-22 year olds approximately 22%	NHS Digital (2023)	High
Mental health referrals more than doubled since 2017	NHS/CAMHS data; YoungMinds; Centre for Mental Health	High, regional variation
UK children among least happy at school in developed nations	OECD PISA (2022); Children’s Society Good Childhood Report (2023)	High
76% of young people cite school as a factor in mental health difficulties	YoungMinds (2023)	High, self-report survey
Test anxiety predicts underperformance and psychological distress	Putwain and Daly (2014)	High
Performance-contingent self-worth predicts anxiety and depression	Putwain et al. (2021)	High, longitudinal
School belonging predicts mental health outcomes	Allen and Kern systematic review (2017)	High, cross-cultural
Excluded students disproportionately have unmet SEND and mental health needs	Timpson Review (2019)	High, government-commissioned
Black Caribbean pupils three times more likely to be permanently excluded	DfE exclusion statistics, confirmed through 2024-25	High, official data
335,700 suspensions in Autumn 2024	DfE statistical release (2024-25)	High, official data
One third of secondary students report bullying annually	Anti-Bullying Alliance (2023)	Moderate, self-report
Bullying victimisation independent risk factor for depression and anxiety	Gini and Pozzoli meta-analysis (2009); UK replication studies	High
Whole-school anti-bullying interventions produce modest real reductions	Gaffney et al. systematic review (2021)	High, variable implementation
Positive teacher-student relationships predict wellbeing outcomes	EEF Social and Emotional Learning guidance (2021)	High
90% of teachers supported a student with mental health problems in past year	National Education Union survey (2023)	High, sector survey
Only 35% of teachers feel adequately trained in mental health support	National Education Union survey (2023)	High, sector survey
Year 7 transition associated with increased anxiety and reduced belonging	Topping (2011) and subsequent UK cohort studies	High
CAMHS referral spike following GCSE results	CAMHS referral pattern data, multiple years	High, administrative data
Some group wellbeing interventions carry iatrogenic risk via peer contagion	Recent systematic reviews; EEF implementation guidance	Moderate, emerging
School refusal and persistent absence predict poor long-term mental health	Multiple longitudinal studies	High
Evidence-based whole-school wellbeing approaches produce measurable outcomes	EEF; What Works Centre for Wellbeing	High, implementation-dependent

References

Allen, K. and Kern, M.L. (2017). School Belonging in Adolescents: Theory, Research and Practice. Springer.

Anti-Bullying Alliance (2023). Annual Bullying Survey. https://anti-bullyingalliance.org.uk

Children’s Society (2023). The Good Childhood Report 2023. https://www.childrenssociety.org.uk/good-childhood

Department for Education (2024-25). Permanent and Fixed-Period Exclusions in England. https://www.gov.uk/government/collections/statistics-exclusions

Education Endowment Foundation (2021). Social and Emotional Learning: Guidance Report. https://educationendowmentfoundation.org.uk

Gaffney, H., Farrington, D.P. and Ttofi, M.M. (2021). Examining the effectiveness of school-bullying intervention programs globally. Educational Psychology Review, 33. https://doi.org/10.1007/s10648-020-09548-4

Gini, G. and Pozzoli, T. (2009). Association between bullying and psychosomatic problems. Pediatrics, 123(3). https://doi.org/10.1542/peds.2008-1215

Keeping Children Safe in Education (2025). Statutory Guidance for Schools and Colleges. https://www.gov.uk/government/publications/keeping-children-safe-in-education–2

National Education Union (2023). Mental Health and Schools Survey. https://neu.org.uk

NHS Digital (2023). Mental Health of Children and Young People in England, 2023. https://digital.nhs.uk/data-and-information/publications/statistical/mental-health-of-children-and-young-people-in-england/2023-wave-4-follow-up

OECD (2022). PISA 2022 Results: Wellbeing at School. https://www.oecd.org/pisa

PSHE Association (2023). Digital Wellbeing Guidance. https://pshe-association.org.uk

Putwain, D.W. and Daly, A.L. (2014). Test anxiety prevalence and gender differences in a sample of English secondary school students. British Journal of Educational Psychology, 84(3). https://doi.org/10.1111/bjep.12033

Putwain, D.W. et al. (2021). Achievement emotions and academic achievement: reciprocal relations and the moderating influence of academic buoyancy. Journal of Educational Psychology, 113(4). https://doi.org/10.1037/edu0000636

Timpson Review (2019). Timpson Review of School Exclusion. https://www.gov.uk/government/publications/timpson-review-of-school-exclusion

Topping, K.J. (2011). Primary-secondary transition. Educational Psychology in Practice, 27(3). https://doi.org/10.1080/02667363.2011.595198

What Works Centre for Wellbeing (2022). School Approaches to Wellbeing: Evidence Review. https://whatworkswellbeing.org

YoungMinds (2023). Young People’s Mental Health Survey. https://www.youngminds.org.uk