📚 UK School Mental Health Interventions: What the Evaluation Evidence Shows | 英国学校心理健康干预项目评估结果
Since 2017 the United Kingdom has invested heavily in school-based mental health provision, from Mental Health Support Teams to universal classroom curricula. Between 2018 and 2024 dozens of evaluations, randomised controlled trials and economic analyses were published, and the results are far more mixed than the policy rhetoric suggests. Some targeted programmes produce reliable, moderate benefits; many universal programmes produce effects close to zero. This article summarises the main evaluation findings, explains the methods behind them, and shows how A-Level Psychology and Sociology students can use this evidence in extended answers.
自 2017 年以来，英国在学校心理健康服务上投入巨大，从”心理健康支持团队”到面向全体学生的课堂课程。2018 至 2024 年间，大量评估、随机对照试验与经济学分析相继发表，其结果远比政策宣传复杂：部分针对性项目能产生稳定、中等程度的收益，而许多普遍性项目的效应量接近于零。本文将梳理主要评估结论、说明其背后的研究方法，并展示 A-Level 心理学与社会学考生如何把这些证据用于论述题作答。
1. Why School Mental Health Became a Policy Priority | 为什么学校心理健康成为政策重点
In 2017 roughly one in nine children aged 5-16 in England met the threshold for a probable mental disorder. By 2023 the NHS England survey reported that about one in five 8-16 year olds and nearly one in four 17-19 year olds fell into that category. Because children spend around 30 hours a week in school, policymakers treat the classroom as the most efficient point for early identification, low-intensity support and signposting to specialist services.
2017 年，英格兰 5-16 岁儿童中约每 9 人就有 1 人达到”可能精神障碍”的判定标准；到 2023 年，NHS England 的调查显示，8-16 岁群体中约五分之一、17-19 岁群体中近四分之一属于这一类别。由于儿童每周约有 30 小时在校度过，政策制定者把课堂视为早期识别、低强度支持和转介专科服务最有效的切入点。
The 2017 Green Paper Transforming Children and Young People’s Mental Health Provision set out three pillars that still structure provision today: a designated senior mental health lead in every school, Mental Health Support Teams (MHSTs) working across clusters of schools, and a four-week waiting time pilot for NHS specialist services.
2017 年绿皮书《转变儿童与青少年心理健康服务》提出三大支柱，至今仍构成服务体系框架：每所学校指定一名高级心理健康负责人；由跨校集群组成的”心理健康支持团队”（MHSTs）；以及针对 NHS 专科服务的”四周等待时间”试点。
2. The Main UK Programmes and How They Were Evaluated | 英国主要干预项目及其评估方式
The evidence base is not a single study but a patchwork of designs with very different levels of internal validity. Cluster randomised controlled trials (RCTs) randomise whole schools rather than individual pupils, because pupils within a school share teachers, norms and curriculum. Quasi-experimental and pre-post designs are cheaper but cannot rule out maturation, regression to the mean or selection effects.
证据基础并非单一研究，而是由内部效度差异极大的多种设计拼凑而成。整群随机对照试验（RCT）以学校而非学生个体为单位进行随机分配，因为同一学校的学生共享教师、规范与课程。准实验与前后测设计成本更低，但无法排除成熟效应、均值回归或选择效应。
| Programme | 项目 | Design | 设计 | Headline result | 主要结果 |
| Mental Health Support Teams (MHSTs) | National mixed-methods evaluation, pre-post plus comparison | Significant symptom reduction for treated pupils; reduced onward referrals |
| MYRIAD (.b mindfulness) | Cluster RCT, 84 schools, ≈8,400 pupils | No effect on the primary depression-risk outcome |
| Education for Wellbeing | Cluster RCT, 400+ schools, ≈30,000 pupils | Small or null effects on symptoms; some gains in literacy and help-seeking |
| EEF social and emotional learning trials | Multiple cluster RCTs | Mixed; several well-implemented curricula show no symptom impact |
3. Headline Findings from the Mental Health Support Teams Evaluation | 心理健康支持团队评估的主要发现
The national MHST evaluation, run by the Anna Freud Centre with the Department for Education and NHS England, reported that pupils who received one-to-one or group work from Education Mental Health Practitioners showed statistically significant reductions in emotional and behavioural difficulties. Average Strengths and Difficulties Questionnaire (SDQ) total difficulties scores typically fell by around two to three points, a change generally interpreted as clinically meaningful at the individual level.
由安娜·弗洛伊德中心联合教育部与 NHS England 开展的 MHST 全国评估显示，接受教育心理健康从业者一对一或小组干预的学生，其情绪与行为困难出现统计上显著的下降。 strengths and difficulties 问卷（SDQ）总困难分平均下降约 2 至 3 分，这一变化在个体层面通常被视为具有临床意义。
Two further findings matter. First, teams reported that most children were seen within a few weeks rather than the months typical of specialist CAMHS referrals. Second, MHSTs appeared to reduce onward referrals to specialist services in several areas, which is where much of the claimed cost-effectiveness comes from.
另外两点发现同样重要。第一，各团队报告大多数儿童在数周内即获得服务，而转介专科 CAMHS 通常需要数月。第二，在若干地区，MHST 似乎减少了向专科服务的二次转介——声称的成本效益很大一部分正来源于此。
4. Universal Versus Targeted Approaches: Reading the Effect Sizes | 普遍性干预与针对性干预：如何解读效应量
Effect sizes allow comparisons across programmes that use different outcome measures. For a two-group comparison the standardised mean difference is calculated as the difference between group means divided by the pooled standard deviation.
效应量使使用不同结果量表的项目之间可以相互比较。对于两组比较，标准化均值差的计算方式为两组均值之差除以合并标准差。
g = (M₁ − M₂) ÷ SD pooled
By convention, g ≈ 0.2 is small, g ≈ 0.5 moderate and g ≈ 0.8 large. Targeted, indicated interventions delivered to pupils already showing difficulties typically report g ≈ 0.3-0.6. Universal classroom programmes usually report g ≈ 0.05-0.20, and many report confidence intervals that cross zero, meaning no reliable effect.
按惯例，g ≈ 0.2 为小效应，g ≈ 0.5 为中等，g ≈ 0.8 为大效应。针对已出现困难学生的”指示性”干预通常报告 g ≈ 0.3-0.6；而面向全体的课堂课程通常报告 g ≈ 0.05-0.20，许多研究的置信区间跨越零，意味着不存在可靠效应。
A second useful statistic is the number needed to treat (NNT), which expresses how many pupils must receive an intervention for one additional pupil to benefit.
另一个实用指标是”需治疗人数”（NNT），表示需要多少人接受干预才能多出现一名获益者。
NNT = 1 ÷ ARR (absolute risk reduction)
5. Case Study 1: The MYRIAD Trial and School-Based Mindfulness | 案例一：MYRIAD 试验与校园正念课程
MYRIAD was the largest trial of school-based mindfulness in the world: 84 secondary schools and approximately 8,400 pupils aged 11-14 were randomised either to the ten-lesson .b mindfulness curriculum plus booster sessions, or to teaching as usual. Trained teachers delivered the programme, and outcomes were measured at one and two years.
MYRIAD 是全球规模最大的校园正念试验：84 所中学、约 8400 名 11-14 岁学生被随机分配到十课时的 .b 正念课程（含强化课）或常规教学组。课程由受训教师实施，结果在一至两年后测量。
The primary outcome was the risk of developing depressive symptoms. The trial found no significant difference between groups. Secondary analyses suggested that pupils who already had higher levels of difficulty may have gained slightly, while lower-risk pupils gained little, and some analyses raised the possibility of small iatrogenic effects for that group, although this signal was not consistent across measures.
主要结果是抑郁症状的发生风险，试验发现两组之间没有显著差异。二次分析提示，原本困难水平较高的学生可能略有获益，而低风险学生获益甚微；部分分析还提出该群体可能存在轻微”医源性伤害”的可能，但这一信号在不同测量指标间并不一致。
The practical implication is important: a programme that is popular, well-liked by teachers and plausibly beneficial does not automatically produce measurable symptom change when delivered universally.
其现实启示十分重要：一个受欢迎、教师评价良好、理论上似乎有益的课程，在全員普遍实施时并不自动带来可测量的症状改善。
6. Case Study 2: Education for Wellbeing and the EEF Trials | 案例二：”幸福教育”项目与 EEF 系列试验
Education for Wellbeing was the largest randomised evaluation of mental health interventions ever run in English schools, involving more than 400 schools and around 30,000 pupils, and funded by the Department for Education. It tested several universal approaches, including mindfulness-based curricula, a mental health awareness and help-seeking programme, and a peer-support focused approach.
“幸福教育”是英格兰学校有史以来规模最大的心理健康干预随机评估，涉及 400 多所学校、约 3 万名学生，由教育部资助。它检验了多种普遍性方案，包括正念课程、心理健康意识与求助行为课程，以及以同伴支持为核心的方案。
Overall, the programme found limited evidence of symptom reduction. The clearest gains were in intermediate outcomes: mental health literacy, knowledge of where to seek help, and reduced stigma. Several Education Endowment Foundation trials of resilience and CBT-based curricula, such as whole-class cognitive behavioural programmes for Year 8 and 9, similarly found no impact on depression, anxiety or wellbeing, and one reported a small negative effect on wellbeing.
总体而言，该项目几乎没有发现症状减轻的证据。最明确的收益出现在”中间结果”上：心理健康素养、求助渠道知识以及污名化的降低。教育捐赠基金会（EEF）开展的若干韧性与 CBT 课程试验，例如面向 8-9 年级的整班认知行为课程，同样未发现对抑郁、焦虑或幸福感的影响，其中一项还报告了幸福感的轻微负向效应。
7. Social and Emotional Learning: The Strongest Evidence Base | 社会情绪学习：证据最扎实的领域
Not all universal approaches perform equally. Meta-analyses of social and emotional learning (SEL) consistently find small but reliable benefits. Durlak and colleagues reviewed 213 school-based SEL programmes covering more than 270,000 students and reported improvements in social-emotional skills, attitudes, behaviour and academic achievement of roughly 11 percentile points.
并非所有普遍性方案表现都一样。社会情绪学习（SEL）的元分析一致发现小幅但可靠的收益。Durlak 等人回顾了 213 个校园 SEL 项目、覆盖逾 27 万名学生，报告其社会情绪技能、态度、行为与学业成绩约提升 11 个百分位。
A more recent meta-analysis of 424 studies across 53 countries found significant improvements in emotional skills, prosocial behaviour and academic outcomes, alongside reductions in emotional distress and behavioural difficulties, with effects generally in the g ≈ 0.1-0.3 range. Notably, effects tend to be larger when programmes are sequenced, active, focused and explicit, and when they are delivered by trained staff with fidelity.
一项更新的元分析整合 53 个国家、424 项研究，发现情绪技能、亲社会行为与学业成果显著改善，同时情绪困扰与行为困难有所下降，效应量大致在 g ≈ 0.1-0.3 之间。值得注意的是，当课程具备”有序列、重参与、有重点、显性化”特征，且由受过培训的人员忠实实施时，效应往往更大。
WHO guidance therefore recommends school-based SEL as part of a whole-school approach, while noting that effect sizes are modest and cannot substitute for specialist clinical care.
因此，世界卫生组织的指南建议把校园 SEL 纳入全校性方案，同时指出其效应量有限，不能替代专科临床治疗。
8. Implementation, Fidelity and Cost-Effectiveness | 实施、忠实度与成本效益
Evaluations repeatedly show that implementation quality explains more variance in outcomes than the theoretical model of the programme. When teachers are given insufficient training time, when lessons are squeezed by exam pressure, or when a programme is treated as an extra rather than embedded in the timetable, outcomes approach zero.
多项评估反复显示，实施质量对结果差异的解释力高于项目的理论模型。当教师培训时间不足、课时被考试压力挤压，或项目被视为额外负担而非融入课表时，结果便趋近于零。
Cost-effectiveness analysis uses the cost per quality-adjusted life year (QALY) and compares it with the NICE willingness-to-pay threshold of roughly £20,000-£30,000 per QALY.
成本效益分析使用每获得一个”质量调整生命年”（QALY）的成本，并与 NICE 约每 QALY 2 万至 3 万英镑的支付意愿阈值比较。
Cost per QALY = Total programme cost ÷ QALYs gained
Targeted MHST work for pupils with identified difficulties appears more likely to fall within that threshold than universal curriculum delivery, because the benefits are concentrated in the pupils who receive the intervention. Universal programmes, by contrast, spread a very small average benefit across every pupil, which raises the cost per unit of benefit.
针对已识别困难学生的 MHST 定向服务，比面向全体的课堂课程更可能落在该阈值之内，因为收益集中在接受干预的学生身上。相反，普遍性方案把极
Published by TutorHao | Exam Prep Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导