Evidence for Targeted and Universal Secondary School-Based Programs for Anxiety and Depression: An Overview of Systematic Reviews

Isabel Zbukvic; Samuel McKay; Samantha Cooke; Rebekah Anderson; Vita Pilkington; Lauren McGillivray; Alan Bailey; Rosemary Purcell; Michelle Tye

doi:10.1007/s40894-023-00211-1

Evidence for Targeted and Universal Secondary School-Based Programs for Anxiety and Depression: An Overview of Systematic Reviews

Zbukvic, Isabel; McKay, Samuel; Cooke, Samantha; Anderson, Rebekah; Pilkington, Vita; McGillivray, Lauren; Bailey, Alan; Purcell, Rosemary; Tye, Michelle 2024-03-01 00:00:00 While there are a number of systematic reviews on school-based mental health programs, there appears to be heterogeneity in their overall findings and conclusions, possibly due to the tendency to combine evidence from primary school and high school programs. To investigate the evidence for the effectiveness of targeted (for specific groups) and universal (for all students) mental health programs delivered in secondary schools, a systematic review of systematic reviews was conducted. A systematic search for reviews published from 2015 included outcomes for depression and anxiety—the most common mental health conditions—and quality appraisal of original studies in majority secondary school settings. A total of 395 references were screened and 14 systematic reviews were included. Of reviews that were assessed in full, most were excluded for not clearly separating studies in secondary school settings from primary school settings. Findings were generally positive but heterogenous in terms of quality and long-term outcomes. Interventions were mainly based on cognitive behavioural therapy (CBT), with results showing that targeted interventions are generally effective in the short-term for both anxiety and depression, while universal programs may be effective in some situations, typically in the short term and when programs were CBT-based; where reviews focused on universal resilience programs there were no significant effects. The overview shows a need for systematic reviews focused on secondary school settings, which consider contextual and individual factors that can influence the implementation and effectiveness of programs. Keywords Schools · Students · Adolescent · Humans · Mental Health Introduction in global research and public health campaigns (UNICEF, 2022), as well as the expansion of mental health services Depression and anxiety are amongst the most common men- focused on youth aged 12–25 years (Rocha et al., 2015). tal illnesses affecting adolescents and have been associated While the evidence for early intervention continues to accu- with social, school and family problems, poor physical mulate, there exists a paradox where adolescents experience health, and costs for individuals, families, and communi- some of the highest levels of mental health issues yet have ties (Bitsko et al., 2018). Early intervention and prevention the lowest levels of mental health service access of any age efforts for depression and anxiety aimed at adolescents have group (McGorry & Mei, 2018). This has driven an inter- become a high priority in many parts of the world, reflected est in strategies to support the mental health and wellbeing of young people outside of health systems. One increas- ingly common strategy is the use of school-based mental * Isabel Zbukvic health programs. There have now been several systematic isabel.zbukvic@orygen.org.au reviews published on school-based mental health programs, with mixed conclusions about effectiveness. Most of these Orygen, 35 Poplar Road, Parkville, VIC 3052, Australia 2 reviews, however, have combined research from across pri- Centre for Youth Mental Health, The University mary and secondary school settings, ignoring developmental of Melbourne, Victoria 3010, Australia 3 differences between these age groups as well as contextual Black Dog Institute, University of New South Wales, factors between primary and secondary education settings. Randwick, NSW 2031, Australia Vol.:(0123456789) 1 3 Adolescent Research Review The present overview of reviews seeks to address this gap by There has been much research to date on school-based summarising the range of existing evidence for high school mental health programs and several systematic reviews have based universal and selective mental health programs target- now been published on the topic. However, overall, there ing depression and anxiety. appears to be heterogeneity in their findings and conclu- Secondary school years capture a key developmental sions. This may be due to differences in methodological period for mental health, with global epidemiological data factors such as program type, delivery mode, implementing indicating that half of all lifetime mental disorders begin personnel (e.g., teachers, mental health clinicians, etc.), and by 18 years of age (Solmi et al., 2022). It is increasingly outcomes investigated. Mixed findings may also be because recognised that secondary schools are uniquely placed as many reviews do not separately consider the evidence from settings for effective intervention in youth mental health as primary school and high school programs. Given that the they offer a unique opportunity to reach a large number of prevalence of common mental illnesses increases substan- adolescents and overcome common barriers to help-seeking tially in adolescence as compared to childhood, reviews and and mental health service access such as stigma, physical meta-analyses that conflate the findings from primary and location and cost. Indeed, young people spend more time secondary school programs are ignoring critical develop- at school than in any other institution. The World Health mental considerations that may help to clarify whether and Organization (WHO) acknowledges that schools are well- why school-based programs are effective in different age positioned to support adolescent mental health by provid- groups (Younger, 2016). Overviews of systematic reviews— ing a setting for large-scale implementation of interventions, or, umbrella reviews—involve systematic review and syn- across a range of social contexts (WHO, 2012). In light of thesis of the evidence from multiple, existing systematic this, school-based mental health programs have proliferated reviews in a given area. They are a relatively recent approach in recent years. School-based mental health programs vary to evidence synthesis, and are considered particularly help- widely in terms of intervention type, delivery, the provi- ful for providing insight into the overall state of the evidence sion of ongoing support and involvement of young people. in situations where a number of systematic reviews have For the purposes of the present review, these programs are already been published evaluating interventions for a par- broadly classified into two types: mental health promotion ticular outcome (Hunt et al., 2018). While several systematic programs and mental illness prevention programs. Mental reviews on school based mental health programs targeting health promotion programs are those which focus on positive depression and anxiety have now been conducted, the pre- development and wellbeing, such as social inclusion, social sent review represents the first known umbrella review of and emotional skills, and problem-solving. Mental illness this area. prevention programs are those which focus on preventing the development of common mental disorders, such as anxiety and depression. This review will focus on prevention pro- Current Study grams for depression and anxiety, due to the prevalence of these conditions amongst adolescents and their far-reaching The existing evidence from systematic reviews presents and significant consequences. mixed findings for the effectiveness of school based mental School-based prevention programs can be classified into health programs addressing anxiety and depression, possibly two main approaches: universal approaches and targeted due to combining findings across primary and secondary approaches (Zbukvic et al., 2020). Universal approaches school populations. To address this, the current study aimed involve all students, regardless of their experience, identity, to provide an overview of systematic reviews of research or level of risk for developing a mental disorder. Targeted specifically in secondary schools. The present overview “selective” approaches are aimed at particular groups of identified the overall evidence for different program types, young people who are identified as having a higher risk of strengths and limitations of the evidence, and opportunities developing a mental disorder—for example, students who for future research. The research was guided by the overarch- live with parents experiencing mental ill-health, or students ing research question: What is the evidence for secondary with a neurodevelopmental condition. Targeted “indicated” school-based programs for improving depression and anxi- approaches are aimed at students who are already experi- ety outcomes in young people (aged 12–18 years)? encing identifiable symptoms below the threshold typically associated with a diagnosable mental disorder. The current overview will present the evidence from systematic reviews Methods focused on universal and/or targeted prevention programs in secondary schools, to help build knowledge about which This overview is presented using the preferred reporting program type may be most effective for addressing anxiety items for overviews of reviews (PRIOR) (Gates et al., 2022). and depression in adolescents. An internal protocol was developed to document the search 1 3 Adolescent Research Review strategy, eligibility criteria, data extraction and synthesis search strategy). The Evidence Finder’s anxiety and depres- procedures. The search strategy was pragmatically designed sion search retrievals were searched using the following to originally produce an “Evidence Summary” from key terms in title, abstract, keyword, and label fields: school findings (Zbukvic et al., 2020). The review was not reg- AND (systematic review OR (metaanalys OR meta-analys istered, but the published “Evidence Summary” provides OR meta analys)). Combined, the searching procedure pro- a record of the purpose and key methods used to produce vides coverage for records published from 1980 through to the review. For the present article, only systematic reviews 30 June 2022. were included; no supplementary primary studies or grey Titles and abstracts of retrieved records were then literature were searched or obtained, and all included articles searched in EndNote using the terms “anxiety” or “depres- were assessed for quality and risk of bias. Data collection sion”. Retrieved articles were screened for eligibility by forms and data extracted from the included studies can be abstract, using the eligibility criteria. Abstract and full text provided by the corresponding author on request. screening of articles was conducted by IZ, RA, SC, LM, VP and MT working independently using the eligibility criteria, Search Strategy with final eligibility for inclusion confirmed through discus- sion at team meetings. Where there was any hesitation or Literature was identified using the “Evidence Finder” disagreement about whether a review should be included, (https:// www. orygen. org. au/ Tr ain ing/ Evide nce- F inder), a this was resolved via discussion. publicly available living database and search engine jointly developed by Orygen and headspace (National Youth Mental Eligibility Criteria Health Foundation). The “Evidence Finder” is a comprehen- sive repository of peer-reviewed systematic reviews and con- Inclusion criteria were: systematic reviews of controlled trolled trials evaluating prevention and treatment approaches trials published in English focused on school-based inter- for common mental ill-health conditions with a peak onset ventions targeting prevention of, or reduction in, depression during adolescence/early adulthood (12–25 years). These and anxiety symptoms, with at least half the participant conditions include anxiety, depression, bipolar, psychosis, population (i.e., program recipients) comprising secondary eating disorders, substance-use, and suicide/self-harm. The school-aged young people (i.e., 12- to 18-year-olds). The co- systematic searching and screening methodology used to primary outcomes were anxiety and depression symptoms create and maintain the Evidence Finder is detailed in pub- and could be measured using any tool and at any time-point. lished studies (De Silva et al., 2018; Hetrick et al., 2018). School-based interventions were defined as programs deliv - Reproducible searches are run annually in MEDLINE, ered in secondary school settings, either in-curriculum or Embase and PsycINFO databases. Retrieved records are outside of usual classes (regardless of whether school sta ff or screened against pre-defined eligibility criteria, and included others delivered the intervention). No restriction was placed studies are coded within the database to support searching on intervention type (e.g., psychosocial, lifestyle interven- according to condition or disorder category, stage of illness, tions) or on comparison condition. Systematic reviews were intervention type, and study design. By July 2022, more than defined as using a systematic search strategy with or with- 480,000 articles had been retrieved and screened, and over out meta-analysis. Reviews were required to be published 5000 unique studies included in the Evidence Finder data- from 2015 onwards to focus on recent evidence and limit base, covering publication dates 1980 to 30 June 2021. The primary trial overlap, based on the assumption that literature “Evidence Finder” tool has been used as a searchable infor- included in recent reviews would cover much of the litera- mation source underpinning published systematic reviews, ture included in previously published reviews. scoping reviews and evidence mapping reviews (e.g., Bell Exclusion criteria was based on (i) evidence type (articles et al., 2022; Pascoe et al., 2021). were excluded if they were non-systematic reviews, primary For the present study, the team responsible for maintain- quantitative research studies, policy evaluations, editorials, ing the Evidence Finder (SC, AB) conducted searches in the conference papers, protocols, theses and book chapters), (ii) database for records with study design coded as “systematic participants (reviews were excluded if the mean age of par- review” AND “school” in the title, abstract, keyword or label ticipants was outside of 12–18 years or if less than half the fields. The original search was run in February 2019 and studies were identifiable as being conducted with secondary updated in July 2021 and again in August 2022, identifying school age young people age 12–18 years) (iii) interventions records published to 30 June 2021. To ensure coverage of (reviews were excluded if they described studies of inter- more recently published literature an additional search was ventions that did not target depression, anxiety or mental conducted in August 2022 of unscreened records retrieved health and/or if they less than half the studies were delivered for the Evidence Finder annual update for the period 1 July in a school-based setting) and (iv) outcomes (reviews were 2021 to 30 June 2022 (see Appendix for Evidence Finder excluded if primary outcomes of studies included alcohol 1 3 Adolescent Research Review or other drug use, psychosis, suicidality, physical health or trials (including any multiple counts) (N), and the number academic outcomes). Reviews were also excluded if they of systematic reviews (c) in the overview. Each “index” described interventions delivered in specialist secondary trial represents the first time a primary trial is included in schools. Where a review was later updated, the original an overview. CCA overlap scores are interpreted as slight review was excluded from results. (0–5), moderate (6–10), high (11–15) or very high (> 15). The CCA formula is as follows: Data Extraction and Synthesis N −r CCA = rc − r Six authors extracted data from included publications (IZ, LM, MT, RA, SC, VP) using a structured template spread- The AMSTAR 2 checklist was used to evaluate the qual- sheet. Extracted data included: author, year of publica- ity of included systematic reviews (Shea et al., 2017). The tion, title of publication, country/countries where primary checklist includes 16 items assessing methodological qual- research was conducted, aims, methodology, findings, con- ity. All SRs were assessed by one reviewer (LM, MT or VP) clusions, caveats and biases, and other comments. Results and a second reviewer (SM) duplicated the appraisal of 20% were collated into tables and synthesized into written sum- of SRs, with 93% agreement and only minor disagreements maries based on the approach to prevention (targeted, uni- that had no impact on cond fi ence grades. It is recommended versal), with results classified as statistically significant/ that review authors determine which of 16-items from the non-statistically significant and further by standardized AMSTAR 2 checklist are critical and non-critical for their effect size and 95% confidence interval (CI) or credible review and use these to grade included systematic reviews interval (CrI). Effect sizes were further described accord- (Shea et al., 2017). Based on this recommendation grades of ing to results of included articles. Results were classified confidence for each review were produced based on critical by time-point (immediately post-intervention or follow-up) flaws and non-critical weaknesses (see Table 3). and intervention type where relevant. A table providing an The checklist includes date of publication, to ensure overview of the characteristics of all included reviews was results are up to date. Systematic reviews published in the produced, including details of participants, interventions, last five years (e.g., 2017 onwards) were considered up to analysis and reported results (Table 1). A summary table was date. The final item is the match between the scope of the produced showing the proportion of the statistically signifi- overview and the primary trials within included systematic cant positive effects compared to null effects by intervention reviews. In this overview, only systematic reviews with a type and time-point (Table 2). The time-points for this sum- minimum of 50% of their primary trials fulfilling inclusion mary information were set to immediately post intervention, criteria for age/setting (secondary school) were included. 0–6 months post intervention, 6–12 post intervention and Systematic reviews with a higher proportion of primary 12 + months post intervention. studies fulfilling the inclusion criteria were considered a better match. Risk of Bias Risk of bias summaries across the primary trials included Results in each systematic review was extracted by LM, MT, SC and VP. An established four-item checklist for overviews Results of the Search of systematic reviews was used to assess potential bias both within and across included systematic reviews (Ballard & The original search (February 2019) retrieved 145 reviews, Montgomery, 2017). The four items include: (1) overlap, of which six were included (Bastounis, 2016; Caldwell et al., (2) rating of confidence from the AMSTAR 2 checklist, (3) 2019; Dray, 2017; Ferreira-Vorkapic, 2015; Sancassiani, date of publication, and (4) match between the scope of the 2015; Werner-Seidler, 2017). The updated searches (July included review and the overview itself. Each of these items 2021, August 2022) resulted in the retrieval and screening are described in detail below. of an additional 37 reviews and the inclusion of an additional It is also important to assess overlap in the use of the four articles. A supplementary search (August 2022) within primary studies across multiple systematic reviews, as high the unscreened search retrievals identified for the 2021–2022 overlap can contribute to biased results. The corrected cov- Evidence Finder update resulted in retrieval and screening of ered area (CCA) is a comprehensive and validated meas- an additional 231 reviews and inclusion of an additional five ure of overlap that has been used in a previous review of articles (Bradshaw et al., 2021; Fulambarkar et al., 2022; reviews of school-based interventions (Levinson et al., Kambara & Kira, 2021; Karukivi et al., 2021; Werner- 2019). The CCA is calculated based on three variables: the Seidler et al., 2021). Twenty-five reviews were assessed in number of “index” primary trials (r), the number of total full but not included because the age group did not meet 1 3 Adolescent Research Review 1 3 Table 1 Summary of characteristics and findings of included reviews First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity author ies (total in secondary evidence secondary preven- tion types parison frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) ticipants) or age (%) setting in > 50% Universal Targeted Universal Targeted studies) Brad- 12 studies 6 (50%) System- No pooled Universal Universal No inter- Unable to deter- N/A Unable to deter- N/A Considerable N/A shaw (15,893 atic analysis school- vention, mine pooled mine pooled heterogeneity (2021) partici- review based men- waitlist, efficacy efficacy in measures pants) tal health standard used to intervention cur- assess (CBT, riculum, depression positive active and anxiety psychol- control ogy, health pro- promotion gram, education, placebo social and control emotional wellbeing programs, life skills, other locally developed interven- tions) Bastou- 9 studies 6 (67%) Meta- No Universal Resilience Active Non-significant N/A Non-significant N/A Heterogeneity Sensitivity analysis nis (4,744 analy- programs control, effect effect for anxiety revealed age of (2016) partici- sis (CBT*, no inter- (immediate): (immediate): interven- participants as pants) IPT) vention, SMD = 0.13, MD = − 0.23, tion effects possible source of waitlist 95& CI 95% CI (I = 20%) heterogeneity [0–0.26] [− 1.09–0.62] and Non-significant depression effect on interven- depression: tion effects MD = 0.09, 95% (I = 65%) CI [− 0.61–0.80]; heterogeneity was lower after post- hoc sensitivity analysis, limiting analysis only to studies targeting the same age group, I2 = 36%, Tau2 = 0.25, df = 6, p = 0.15 Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Caldwell 137 studies 85 (62%) Meta- Yes—separate Universal Any universal No inter- Evidence for Evidence for exer- No evidence No evidence for Moderate het- Sensitivity analysis (2019) (56,620 analy- analyses for and or targeted vention, mindfulness cise from single (immediate) any intervention erogeneity conducted by partici- sis secondary targeted program waitlist, and relaxation trial (immediate; Weak evi- (6-12 m) removing studies pants) school age/ (CBT*, IPT, usual (immediate; (SMD − 0·47, dence for Evidence for CBT at high or unclear setting psycho- cur- SMD − 0·65, 95% CrI − 0·86 CBT-based from single risk of bias for supportive riculum, CrI − 1·14 to to − 0·09) interventions study (13-24 m; randomisation programs) attention − 0·19;) No evidence (6-12 m)— SMD − 0·50, and alloca- controls Weak evidence (6-12 m) third-wave 95% CrI − 0·96 tion conceal- for CBT Evidence for CBT CBT (SMD to − 0·05) ment; findings (immediate; from single study − 0·13, 95% unchanged SMD − 0·15, (13-24 m; SMD CrI − 0·27 CrI − 0·34 to − 0·26, 95% CrI to 0·01) and 0·04) − 0·52 to − 0·01) CBT + IPT No evidence (SMD − 0·10, (6–12 m) CrI − 0·26 to No evidence 0·05) (13–24 m) No evidence (13–24 m) Dray 57 studies 38 (67%) Meta- Yes—Meta- Universal Resilience Control Adolescent N/A Adolescent N/A Heterogeneity Not reported analysis programs (not trials: Non- trials: Non- for anxiety (2017) (41,521 analy- partici- sis possible for (CBT*, speci- significant significant interven- 11–18 year- Penn fied) or effects vs con- effects overall tion effects pants) olds Resiliency alternate trol (0-54 m): (0-54 m): (I = 84%) (‘adolescent Program, resil- SMD 0.02, SMD 0.05, and trials’) positive ience- 95% CrI –0.24 95% CrI depression psychology, focused to 0.20 or − 0.11 to 0.01 interven- mindful- inter- vs alternate or vs alternate tion effects ness) vention intervention: intervention: (I = 56%) SMD 0.12, CI SMD 0.08, CI -0.02 to 0.25 -0.08 to 0.24 Ferreira- 9 studies 7 (78%) Meta- No Universal Yoga-based Waitlist, Significant N/A Significant N/A Heterogeneity Not reported for anxiety Vork- (1,144 analy- programs physical effects effects apic partici- sis educa- (immediate): (immediate): interven- tion effects (2015) pants) tion, no SMD − 0·036, SMD − 0·42, inter- 95% CrI 95% CrI (I = 79%) − 0.76 to vention − 0.71 to and − 0.01 − 0.07 depression interven- tion effects (I = 17%) Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Feiss 42 studies 42 (100%) Meta- N/A Universal Any universal Waitlist, Significant effects (imme- Significant effects (immedi- High het- Not reported (2019) (7,310 analy- and or targeted usual diate): t(54) = − 3.72, ate): t(116) = − 3.120, erogeneity partici- sis targeted program care, p < 0.001, dexp = − 0.70, p < 0.01, dexp = − 0.62, for anxiety pants) (indi- (CBT*, attention 95%CIexp: − 0.94, − 0.46, 95%CIexp: − 0.81, − 0.43, interven- cated; stress support, dctrl = − 0.14, 95%CIc- dctrl = − 0.22, 95%CIc- tion effects selective inocula- inactive trl − 0.26, − 0.01 trl: − 0.34, − 0.10 I = 89.26%) by eth- tion*, medi- control, No significant follow-up effects (3-6 m): No significant follow-up effects and moder- nicity, tation, other active t(20) = − 0.72, p = 0.48, dexp = − 1.0, (3-8 m): t(78) = − 0.009, ate to high location, holistic control 95%CIexp: − 1.29 − 0.71, p = 0.99, dexp = − 0.56, among con- income) interven- (e.g. dctrl = − 0.77, 95%CIctrl: − 0.76, 0.48 95%CIexp − 0.81, − 0.31, trol effects tions wellness dctrl = − 0.56, 95%CIc- (I = 63.24%) course) trl: − 0.89, − 0.22 High heteroge- neity among depression interven- tion effects (I = 96.91%) and control effects (I = 95.07%) Fulam- 9 studies 9 (100%) Meta- N/A Universal Mindfulness- Usual cur- No significant N/A No significant N/A Studies were Not reported barkar (5,046 analy- based inter- ricula, effects overall effects overall conducted (2022) par- sis ventions waitlist, (0-11wks, (0-11wks, in varied ticipants) active CI 95% CI 95% settings, Note: control − .05,− .44) − .14,− .53) included a 1 study (e.g. range of par- (N = 145) yoga, ticipants e.g., examined relaxa- clinical and stress not tion, non-clinical anxiety/ health samples, and depres- cur- used various sion riculum, interventions outcomes sub- (Anxiety stance homogeneity use χ = 13.01, preven- p < .01 tion) Depression homogeneity χ = 29.91, p < .01) Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Gee 45 studies 36 (80%) Meta- No Targeted Manualised Passive N/A Significant effect N/A Significant Overall Excluding data (2020) (number analy- (indi- psycho- controls (immediate): effect (0-6 m): heterogeneity from three of partici- sis cated) logical (waitlist, SMD = − .49, 95% SMD = .34, 95% I = 81% depression- pants not programs usual CI − 0.79, − 0.19 CI − 0.48, − 0.21 focused cluster reported) (CBT or care, No significant No signifi- randomised trials, CBT- moni- follow-up effects cant effects the effect size for based* toring (0—> 12 m): (6—> 12 m): depression symp- includ- symp- SMD = -.45, 95% SMD = -.10, toms at postinter- ing Penn toms, CI − 1.34, 0.43 and 95% CI − 0.35, vention remained Resiliency psych- SMD = -.06, 95% 0.14 and in the small– Program, oeduca- CI − 0.26, 0.13 SMD = -.10, medium range IPT or IPT- tion), 95% CI − 0.28, (SMD = − .39, based, other self- 0.08 95% purely psy- help, CI − 0.54, − 0.25, chological indi- p < .001, k = 28) programs vidual support, active controls (not defined) Hugh- 20 studies 12 (60%) Meta- No Targeted Any universal No inter- N/A Significant effects N/A N/A Heterogeneity Studies with a low Jones (2076 analy- (indi- prevention vention, (0-12 m): I = 78% risk of contami- (2021) partici- sis cated) or early waitlist, g = − 0.28, 95% nation had an ES pants) intervention attention CI = − 0.50, − 0.05 of g = 0.03 (k = 5, program control and 12 m: 95% CI = − 0.14, (CBT*, play g = − 0.24, 95% 0.21) therapy, CI = − 0.48, 0.00 mindful- No significant ness, follow-up emotion effects (> 12 m): regulation, g = − 0.01, 95% working CI = − 0.38, 0.36 memory) Kambara 18 studies 18 (100%) Meta- N/A Universal Universal or No inter- N/A N/A No signifi- Significant effects Moderate Not reported cant effects (immediate): heterogeneity (2021) (3121 analy- and targeted vention, partici- sis targeted intervention waitlist, (immediate): g = − 0.56, in pairwise (g = − 0.04, 95%Cl comparison pants) (indi- programs usual cated) based on care, 95%Cl [− 0.78, − 0.34] for universal usual [− 0.27, 0.18] (I CBT* = 53.72%) cur- and selective riculum, (I = 52.12%) attention No significant follow-up effects Moderate Not reported control (3–12 + months) combined heterogeneity universal and selective approaches: (I = 51.58%) g = − 0.12, 95%Cl [− 0.30, 0.06] Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Karukivi 8 studies 8 (100%) System- N/A Targeted Targeted Placebo N/A Significant effects for N/A Significant effects Not reported N/A (2021) (425 atic (indi- interven- control, CBT (immediate, for CBT, partici- review cated) tions active 0–12 m) attention-bias- pants) (CBT*, control, No statistical analy- modification inter- no inter- ses performed for (immediate) personal vention, motivation-adap- 0–12 m) counselling, no tive-skills-trauma No significant attention- control resolution effects for bias modi- interpersonal fication, counselling motivation- (immediate, adaptive- 0-6 m) skills- No statistical anal- trauma yses performed resolution for motivation- adaptive- skills-trauma resolution Ma 38 studies 30 (79%) Meta- No Universal Resilience- Waitlist, N/A N/A Significant Significant effect Heterogeneity Not reported (2020) (24,135 analy- and oriented assess- effect (immediate): = 79.2% partici- sis targeted CBT inter- ment (immediate): g = 0.21, 95%Cl pants) (selec- ventions* only, g = 0.09, [ 0.06, 0.35] tive attention 95%Cl [ 0.02, And 0-6 m: g = 0.24, 95%Cl e.g. “at control, 0.16] No risk”; no inter- significant [ 0.10, 0.38] indi- vention, follow-up cated) placebo effects (0-6 m): g = 0.05, 95%Cl [- 0.03, 0.13] Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Sancas- 22 studies 13 (59%) System- No pooled Universal Emotional and Usual cur- No signifi- N/A No signifi- N/A Findings N/A siani (49,169 atic analysis social skills riculum, cant effects cant effects reported as (2015) partici- review programs no inter- (0–6 m) (0–6 m) very hetero- pants) (life skills vention, geneous for training, treat- age/grade of wellbeing ment as participants, programs) usual character- (not istics of speci- intervention fied), and control minimal groups, contact duration of (e.g. programs mailed and follow- reading up, outcomes materi- and tools als), physical educa- tion, other educa- tion or program Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Werner- 130 studies 90 (53%) Meta- Subgroup Univer- Psychological Waitlist, Effect sizes small Effect sizes small Moderate for Not reported Seidler (45,924 31% analy- analy- sal or or psychoe- no inter- (immediate; n = 72, g = 0.18, 95%CI: (immediate; n3 = 101, g = 0.21, both depres- (2021) partici- 10–14 years sis sis—no targeted ducational vention 0.12–0.26) 95%CI: 0.17–0.24)sion (I = 47, pants) (‘early ado- difference (indi- program e.g. (short-term; n = 29, g = 0.19, 95%CI: (short-term; n = 60, g = 0.17, 95%CI: 95%CI: lescent’) between cated; (CBT or usual 0.12–0.26) 0.13–0.22) 0.33–0.58) 45% size of selec- CBT- cur- (medium-term; n = 31, g = 0.23, 95% (medium-term; n = 48, g = 0.10, and anxiety 14–19 years effects by tive by based*, riculum, CI: 0.10–0.35) 95%CI: 0.06–0.13) (I = 46, (‘older ado- age “risk” IPT, mind- active (long-term; n = 8, g = 0.11, 95%CI: (long-term; n = 20, g = 0.10, 95%CI: 95%CI: lescents’) e.g. fulness, control 0.03–0.18) 0.05–0.15) 0.29–0.59) parental personality- (not No significant No significant differ - Significantly Significantly larger psycho- focus, other speci- differences at ences at immedi- smaller effect size com- pathol- therapeutic fied) immediate, ate, short, medium effect size pared to univer- ogy, approaches) short, medium or long-term compared sal (immediate; person- or long-term follow up between to targeted n = 43 g = 0.29, ality, follow up universal and tar- (immedi- 95%CI: behav- between geted programs ate; n = 58, 0.22–0.37) ioural universal g = 0.17, No significant dif- prob- and targeted 95%CI: ferences at short, lems) programs 0.13–0.21) medium or long- No significant term follow up differences between univer- at short, sal and. targeted programs medium or long-term follow up between universal and. targeted programs CBT cognitive behavioural therapy, IPT interpersonal therapy indicates majority (> 50%) of programs. Where specific interventions are included in results columns, this reflects where significant findings were only reported for those interventions Adolescent Research Review Table 2 Summary of review findings by outcome, approach, and effectiveness over time Anxiety Depression Universal Targeted Universal Targeted n % CCA n % CCA n % CCA n % CCA Immediate positive effects 6 66.7 (60.0) 12.7 5 100.0 (100.0) 12.9 5 45.5 (45.5) 9.3 6 100.0 (85.7) 12.0 Immediate no effect 4 44.4 (40.0) 4.3 0 0.0 (0.0) NA 6 54.5 (54.5) 4.6 1 16.7 (14.3) NA 0–6 months positive effect 2 22.2 (28.6) 9.8 2 40.0 (40.0) 0.0 2 18.2 (22.2) 37.4 3 50.0 (42.9) 10.0 0–6 months no effect 5 55.6 (71.4) 4.2 3 60.0 (60.0) 11.8 7 63.6 (77.8) 5.8 4 66.7 (57.1) 5.5 6–12 months positive effect 2 22.2 (50.0) 9.8 2 40.0 (50.0) 0.0 2 18.2 (40.0) 37.4 2 33.3 (50.0) 0.0 6–12 months no effect 2 22.2 (50.0) 13.7 2 40.0 (50.0) 13.8 3 27.3 (60.0) 2.1 2 33.3 (50.0) 4.1 > 12 months positive effect 1 11.1 (25.0) NA 2 40.0 (66.7) 37.4 1 9.1 (20.0) NA 2 33.3 (66.7) 37.4 > 12 months no effect 3 33.3 (75.0) 9.4 1 20.0 (33.3) NA 4 36.4 (80.0) 8.4 1 16.7 (33.3) NA Number of SRs on topic 9 5 11 6 All results positive effect 40.9 64.2 31.9 61.3 All results no effect 59.1 35.8 68.1 38.7 Percentages show the proportion of reviews reporting the relevant effect out of the total number of reviews on that topic. Percentages in paren- theses are adjusted for the number of relevant effect sizes at that time point, as some reviews reported multiple effects for a single time point when comparing different types of interventions. Positive effects include any statistically significant positive effects regardless of effect size. n the number of relevant effects reported by reviews for each effect type. CCA Corrected Covered Area for the specific effect calculated based on the number of reviews and associated included primary studies reporting that effect. NA Not Applicable and is shown where it was not possible to calculate the CCA due to only one or less reviews reporting the relevant effect Adjusted percentage based on overall number of effects reported across all time points Table 3 AMSTAR 3 checklist quality appraisal of included studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Overall score Bastounis (2016) + + + + + + + + + − + + + + − − Moderate quality Bradshaw (2021) + + + + + + + + + − NMA NMA + + NMA − Moderate quality Caldwell (2019) + ± + ± + + + + + − + + + + + + High quality Dray (2017) + + + ± + + + + + − + + + + + + High quality Feiss (2019) + + + ± + + + + + − + − + + − + Moderate quality Ferreira-Vorkapic (2015) + − + ± + + + ± ± − NMA NMA + + NMA + Moderate quality Fulambarkar (2022) + − − ± + − + + + − + − − − − − Critically-low quality Gee (2020) + + + ± + + + ± ± − + − + + + + Moderate quality Hugh-Jones (2021) + + + + + + + + + − + + + + + + High quality Kambara (2021) + − + + + + + + + − + ± + + + + Moderate quality Karukivi (2021) + − + ± + − ± ± + − NMA NMA + − NMA + Critically-low quality Ma (2020) + − − ± − + + + + − + − + + + + Low quality Sancassiani (2015) + − + − − − − + - − NMA NMA + + NMA + Critically-low quality Werner-Seidler (2021) + + + + + + + + + − + + + + + + Moderate quality High Quality = Zero or one non-critical weakness. Moderate quality = More than one non-critical weakness. Low quality = One critical flaw with or without non-critical weaknesses. Critically low = More than one critical flaw with or without non-critical weaknesses. NMA = Not a meta- analysis. Item 1 in the checklist refers to: Did the research questions and inclusion criteria for the review include the components of PICO? criteria (Cordier et al., 2021), the settings of included studies include depression or anxiety (Bennett, 2015; Das, 2016; were not majority secondary school (Bennett, 2015; Cristea, Labelle, 2015; Lee et al., 2017; Meekums, 2015; O'Dea, 2015; Erford, 2015; James, 2015; Labelle, 2015; Oldham- 2015; Weaver, 2015; Zalsman, 2016), studies were not Cooper, 2017; Townshend, 2016) or could not be confirmed quality-appraised (Bernecker, 2017; Nanthakumar, 2018), as majority secondary school setting (Ahlen, 2015; Hetrick, or the record was an abstract not a review paper article (Cha- 2015; Hetrick, 2016; Stockings, 2016), outcomes did not har, 2015; Cowen, 2014; Galling, 2018). One review was 1 3 Adolescent Research Review updated in the time that searches were conducted so the orig- et al., 2021). Table 1 provides an overview of all included inal review has been excluded from results (Werner-Seidler, reviews with key findings for depression and anxiety out- 2017). Figure 1 provides an overview of the exclusion and comes, number of included studies and participants, pro- inclusion of reviews. gram and intervention types, and comparison conditions. The majority of studies focused on CBT-based interven- Study Characteristics tions, but reviews included research on a range of inter- ventions including health promotion, life and social skills, In total, fourteen systematic reviews were included that resilience programs, yoga-based programs, interpersonal considered findings for school-based mental health pro- therapy-based programs, emotion regulation and other grams with outcomes for anxiety and/or depression psychological and physical health-based interventions. (Bastounis, 2016; Bradshaw et al., 2021; Caldwell et al., Most significant findings related to reviews of majority 2019; Dray, 2017; Feiss et al., 2019; Ferreira-Vorkapic, CBT-based programs. Where reviews focused on univer- 2015; Fulambarkar et al., 2022; Gee et al., 2020; Hugh- sal resilience programs, there were no significant effects. Jones et al., 2021; Kambara & Kira, 2021; Karukivi et al., All reviews included studies with a range of comparison 2021; Ma et al., 2020; Sancassiani, 2015; Werner-Seidler Fig. 1 Flow diagram for inclu- Records identified from systematic Records removed before sion and exclusion of reviews reviews search screening Evidence Finder databse (n=183) Duplicates (n=19) Evidence Finder unscreened search retrievals (n=231) Records screened From systematic reviews search (n=395) Records sought for retrieval From systematic reviews search (n=38) Records assessed for retrieval Records excluded From systematic reviews search (n= From systematic reviews search: 38) Outcomes not depression or anxiety (n=8) Total included Setting not majority secondary school (n=7) Systematic reviews search (n=14) Setting could not be confirmed as majority secondary school (n=4) Not a review paper (n=3) Papers included in review not quality appraised (n=2) Age group did not meet criteria (n=1) 1 3 Adolescent Research Review conditions including no intervention, waitlist, attention reported no effect (Bastounis, 2016; Sancassiani, 2015). Lik - control, and many compared to usual curriculum. wise for anxiety, there was some evidence for effectiveness of universal programs short term (Ferreira-Vorkapic, 2015) Description of Review Findings while others found no effect (Bastounis, 2016; Sancassiani, 2015). There were reasonably consistent positive findings for Although the present overview only included systematic targeted programs from reviews that combined results from reviews of research majority conducted in secondary school primary and secondary settings in the short-term, however settings, most reviews combined findings from across sec- these were not maintained beyond 6–12 months (Gee et al., ondary and primary school participants or settings. Out of 2020; Hugh-Jones et al., 2021; Ma et al., 2020). fourteen reviews, four focused on studies specifically in sec- Overall, there was more evidence available for univer- ondary school settings (Feiss et al., 2019; Fulambarkar et al., sal than targeted interventions, and only a small number of 2022; Kambara & Kira, 2021; Karukivi et al., 2021). Of reviews directly compared these approaches. Table 2 pro- those four, two reviews examined the effectiveness of uni- vides a summary of the overarching findings across reviews versal programs and found no significant effects overall for and shows that evidence was strongest for the efficacy of tar - either anxiety or depression outcomes (Fulambarkar et al., geted interventions for both depression and anxiety, particu- 2022; Kambara & Kira, 2021) while two reviews that exam- larly directly after the intervention, with greater heterogene- ined targeted indicated programs found significant effects for ity in longer term effects. In contrast, universal programs anxiety and depression outcomes for CBT-based programs were found to be effective in less than half of the reviews up to 12 months (Kambara & Kira, 2021; Karukivi et al., on anxiety, and just over one-third for depression, with less 2021) and one review that combined findings from univer - support again for the efficacy of such programs long term. sal and targeted programs found significant effects in the Across reviews, facilitators included health workers and short- but not long-term (Feiss et al., 2019). Three reviews doctors, researchers, school counsellors, teachers, psycholo- conducted separate (Caldwell et al., 2019; Dray, 2017) or gists, graduate students, and other trained instructors. There sub-analyses (Werner-Seidler et al., 2021) with adolescent was no evidence for different outcomes between mental age groups. From these findings, there was weak evidence health professionals or teachers in two reviews (Bastounis, for CBT-based programs as a universal intervention for anxi- 2016; Caldwell et al., 2019), while another review found that ety and depression up to 12 months, and no evidence for interventions were more effective, especially at longer term effectiveness beyond a year (Caldwell et al., 2019), while follow ups, when delivered by mental health professionals another review found no significant evidence for universal (Ma et al., 2020). Another review found that indicated inter- programs in secondary settings (Dray, 2017). These reviews ventions delivered by teachers had no impact on depression also produced some limited evidence for CBT-based pro- or anxiety symptoms (Gee et al., 2020). The original and grams as a targeted intervention for both anxiety and depres- updated review published by one group suggested a slight sion, but where results were combined for universal and tar- preference for external personnel over school staff, though geted interventions in a review focused on secondary school, the earlier review found no significant difference between effects were limited to the first 3 months post-intervention groups for anxiety outcomes, and across both reviews’ find- (Feiss et al., 2019). No reviews compared the effectiveness ings were not maintained at longer-term follow-up (Wer- of targeted or universal approaches in primary vs secondary ner-Seidler et al., 2021; Werner-Seidler, 2017). Digital and school settings, however, one review conducted subgroup face-to-face approaches also appear to be similar in terms analysis and found no difference between effect sizes by age of effectiveness, although there is less evidence currently (Werner-Seidler et al., 2021). That review found small effect available for digital interventions (Caldwell et al., 2019; Gee sizes for universal and targeted programs overall across mul- et al., 2020; Werner-Seidler et al., 2021). Most studies in tiple time-points beyond 12 months after intervention. When included reviews were conducted English-speaking coun- compared using subgroup analyses, there were no signifi- tries including Australia, Canada, and the USA. cant differences between universal and targeted programs for anxiety, while for depression there were significantly smaller Quality Appraisal of Included Systematic Reviews effect sizes for universal programs compared to targeted in the short term, though these were not maintained at follow- The CCA was found to be 5.98%, which represents a moder- up (Werner-Seidler et al., 2021). ate overlap between the 14 systematic reviews covering 319 For reviews that combined results from research with studies. Detailed explanations and calculations for the CCA primary and secondary school students without separate can be found in the Appendix. Table 3 presents the checklist analyses, there was mixed evidence for universal programs. for quality appraisal of included reviews. The majority of For depression outcomes, some reported significant effects reviews were of moderate to high quality, although there was (Ferreira-Vorkapic, 2015; Ma et al., 2020) while others one low and three critically-low quality reviews. Of note, 1 3 Adolescent Research Review out of the four reviews that focused on secondary school reviews of research on school-based universal or targeted programs, two were of critically low quality and two moder- interventions for depression or anxiety, with findings from ate quality. A detailed overview of the AMSTAR 2 review generally high-quality reviews suggesting that programs can rating checklist and quality appraisal of primary studies can be effective short-term but benefits may diminish over time. be found in the Appendix. Table 4 shows the target outcome Overall, the results of the present review show that tar- of each included systematic review along with the results geted school-based interventions are generally effective at for other 3 items from the checklist for assessing risk of reducing depression and anxiety symptoms in young peo- bias in overviews of reviews. All but three of the systematic ple, especially at the time of intervention, while universal reviews were considered up to date by AMSTAR criteria of programs show greater heterogeneity in results. Regardless published in the last five years. The proportion of included of intervention or program type, evidence for the long-term relevant primary studies targeting secondary school age efficacy of school-based interventions for depression and young people ranged from 50 to 100%, with four reviews anxiety is still mixed and highly variable. It is not clear including 100% relevant studies. which programs may be most suitable for secondary school students, as most reviews combined findings from across age groups and settings. The combined results provide some Discussion basis for integrating universal or targeted interventions for anxiety and depression into school health and wellbeing pro- There have now been several systematic reviews of school- grams. However, the heterogeneity in findings and quality based mental health programs aimed at anxiety and depres- across the reviews limits strong conclusions. In particular, sion published without clear conclusions for effectiveness, limitations in the participant cohorts, study designs, and pro- possibly due to combining findings from primary and sec- gram implementation strategies leave a number of questions ondary school settings without considering the develop- and opportunities for future investigation. mental differences between age groups, or the heightened period of vulnerability for the emergence of anxiety and Evidence for Different Program and Facilitator Types depression symptoms during secondary school years. This umbrella review sought to address this issue by undertaking The present overview provides a summary of the different a developmentally focused examination of existing evidence program and facilitator types that have been evaluated in for programs delivered in secondary schools specifically. school-based mental health programs. Consistent with pre- The current umbrella review included fourteen systematic vious research, results suggest that targeted programs may Table 4 Risk of Bias assessment for included systematic reviews First Author Target outcome AMSTAR 2 Rating (item 2) Publication % of included studies rel- b c date (item 3) evant by age/setting (item 4) (%) Bastounis Depression Moderate quality 2016 67 Bradshaw Anxiety & depression Moderate quality 2021 50 Caldwell Anxiety & depression High quality 2019 62 Dray Resilience High quality 2017 67 Feiss Anxiety & depression Moderate quality 2019 100 Ferreira-Vorkapic Anxiety & depression Moderate quality 2015 78 Fulambarkar Anxiety, depression & stress Critically low quality 2022 100 Gee Anxiety & depression Moderate quality 2020 80 Kambara Depression Moderate quality 2021 100 Karukivi Anxiety, depression & conduct disorder Critically low quality 2021 100 Hugh-Jones Anxiety High quality 2021 60 Ma Depression Low quality 2020 79 Sancassiani Wellbeing, positive development, healthy Critically low quality 2015 59 lifestyle & academic performance Werner-Seidler Anxiety & depression Moderate quality 2021 76 See appendix for a detailed overview of the AMSTAR 2 rating for each included systematic review All reviews published since 2017 were considered up to date Proportion of studies within the systematic review that clearly met the current study inclusion criteria 1 3 Adolescent Research Review be more effective than universal programs (Werner-Seidler efficiency outcomes, which can build relevant knowledge to et al., 2021). Although it appears that targeted approaches support decision-makers (Moir, 2018). may be more effective than universal, it is essential to acknowledge that the direct comparison of different pro- Developmental Considerations gram types is threatened by confounding factors, particularly symptom levels at baseline. Universal programs include all The present overview aimed to understand the evidence individuals in a cohort, regardless of mental health needs, for secondary school mental health programs by including whereas targeted interventions focus on those who have systematic reviews where it was possible to identify that at greater mental health support needs and therefore have least 50% of primary studies were conducted in a second- greater potential for improvement relative to universal pro- ary school setting or with secondary school age adolescents. grams. Such design characteristics make it difficult to effec- All but one included > 60% studies with secondary school tively compare universal and targeted programs, and may age children, suggesting the majority of primary studies in lead to less positive conclusions about the efficacy of uni- included reviews aligned with the overview scope. How- versal interventions than are warranted (Feiss et al., 2019). ever, the mixture of findings from primary and secondary Overall, while findings from the present overview helps to school studies in most included reviews means that results paint a picture of the evidence for school-based programs should still be interpreted with some caution. While more targeting secondary school students, they also help to show recent reviews (e.g., Werner-Seidler et al., 2017, 2021) have where there are current gaps in the literature and opportuni- addressed this to some extent using sub-group analyses, ties for future research. there is a clear need for systematic reviews that focus on Programs studied in the included reviews were primar- each age group and setting. The present overview offers a ily cognitive behavioural therapy-based. While these pro- step in this direction, presenting the evidence from system- grams appear to be the most effective targeting depression atic reviews of majority secondary school-based research. and anxiety, such conclusions can only be made tentatively The developmental differences between primary school as more research on non-CBT programs is needed. There age (5–11 years) and secondary school age (12–18 years) are was mixed evidence for the impact of facilitators across significant, particularly in relation to mental health preven- the reviews, with most delivered by mental health profes- tion and treatment needs. These age groups show differences sionals or teachers. Taken together, the reviews suggest in cognitive abilities, neurobiology, relationships with peers that interventions may be more effective when delivered by and family; there are also significant differences that come mental health professionals rather than teachers, especially following puberty in terms of romantic relationships, explo- for indicated interventions. Results also suggested similar ration of sexuality and gender, and bodily changes, as well effectiveness for programs delivered in person compared to as new social pressures related to academic performance and online, however the research in this area was limited. At working towards independent living, all of which can impact least one randomized controlled trial is currently underway and be impacted by mental health and wellbeing (National focused on understanding the effectiveness of online app- Academies of Sciences, 2019; Pfeifer & Allen, 2021). These based interventions for mental health delivered in secondary developmental differences have been proposed to explain school settings, which will add to the evidence in this area the rise in rates of mental ill-health between childhood and (Werner-Seidler et al., 2020). From the present findings, it adolescence. It is perhaps not surprising that targeted pro- is not clear under what conditions specific facilitators and grams show stronger evidence for effectiveness in this age delivery methods can have the greatest impact. group compared to findings for universal programs, where Future research designs should consider the need to targeted programs can both address the specific needs of understand how, why, and when secondary school programs adolescents (e.g., peer pressure, academic stress) as well can be most effective. The importance of separating research as being more likely to produce positive impacts on symp- in primary and secondary school settings is relevant here toms if starting from a higher baseline. Results of the pre- too, with major differences in learning environment, pro - sent overview showed that some reviews combining results cesses and structures that can all impact implementation, for primary and secondary school settings reported positive which can in turn influence the reach and impact of pro- results for universal programs, whereas reviews focused grams (Pearson et al., 2015). Although not the focus of the specifically on secondary school did not. It is possible that current review, an understanding of the evidence for cost- universal programs may be more suitable and effective for effectiveness will also help with decision-making about primary school students, which may be confounding results future investment in universal and/or targeted school-based for reviews that combined age groups and settings. This mental health programs. The e fi ld of implementation science may be due to depression and anxiety symptoms being offers frameworks to help capture factors that may influ- more amenable to universal intervention earlier in life, or ence implementation in school settings, as well as cost and it may be related to differences between settings making 1 3 Adolescent Research Review implementation more effective. Future research could help is crucial, as one review found that found that programs to disentangle these questions. The mixed and weaker effects delivered in lower socioeconomic status secondary school for long-term effects for school-based programs overall may settings were less effective than in high or mixed socioeco- also be interpreted through a developmental lens, consider- nomic schools (Caldwell et al., 2019). Similarly, assessing ing the rapid and significant changes that occur for adoles- the effectiveness and acceptability of school-based interven- cent over periods of one year or more. It may be the case tions for depression and anxiety for those from culturally that targeted programs benefit from being delivered regularly and linguistically diverse and LGBTQI + communities, is across secondary school years, however, further research is vital. Such groups often face unique stressors and barriers needed to support this approach. to help seeking that can exacerbate mental health problems, and may require tailored or adapted interventions to best Limitations of Overview Review Methodology support their needs (Brown et al., 2016). Future research and systematic reviews must consider the contextual and Overviews of systematic reviews can provide a comprehen- individual factors that can influence the implementation and sive summary of the existing evidence on a topic, but also effectiveness of school-based programs. Those responsible come with limitations (Ballard & Montgomery, 2017). Over- for program design, implementation, and evaluation should views are unlikely to include the most recent primary studies aim to capture inclusive data about student identity and due to the time lag between study publication and inclusion background, and to assess programs delivered in a range of in systematic reviews. Therefore, it is possible that more settings. Where the evidence is limited, designers and facili- recent studies have addressed the identified research gaps in tators should adapt school-based mental health programs this study. Additionally, scope mismatch, or when the scope according to relevant guidelines and models to ensure cultur- of included systematic reviews is not directly aligned with ally responsive practice (Yohannan & Carlson, 2019). the overview, can also pose a problem, as conclusions may Another limitation of included research relates to study be based on studies that do not fully address the scope of design and outcome measures. While long-term outcomes the overview. Finally, overlap of primary studies between were mixed overall, interpretation is somewhat limited by included systematic reviews can bias results and conclu- long-term impacts not being consistently measured by the sions of overviews. The current study used a CCA analysis, same tools. The development of meaningful, appropriate demonstrating a moderate level of overlap in primary stud- youth mental health measures and outcomes is an area of ies. Critically, overviews can also only provide conclusions research focus that will help with future school-based men- that are accurate when the included reviews are of a high tal health program evaluation and research (Thapa Bajgain standard. Four of the reviews in this overview were critically et al., 2023). Consensus on measures and study designs will low- or low-quality confidence based on the AMSTAR 2 help to promote consistency across the field and enhance checklist, which included two of the four reviews focused on opportunities to identify the key mechanisms that could be secondary school research, with the rest of either moderate leveraged to improve program delivery and efficacy. Con - (n = 7) or high confidence (n = 3). Given the stringency of the sistency in the measures used to assess mental health out- checklist, this suggests that overall quality of the included comes would also support comparison between approach reviews is relatively good (Levinson et al., 2019), however and program types. Without such information, policy mak- the poor quality found for secondary school research casts ers, researchers, and leaders in school health are limited by some doubt on the conclusions that can be drawn from those the evidence-base in terms of recommendations for imple- reviews. Results of the present overview must be considered mentation. Most studies compared programs with “non- in light of these limitations and strengths. The present study active” comparison conditions, for instance where students is among the first umbrella reviews focused on adolescent experience the usual curriculum. This makes it difficult to research, summarizing the state of knowledge in the area of determine in this review whether positive effects relate to secondary school-based mental health interventions (Lev- the components of programs thought to be beneficial for esque, 2016). mental health, such as cognitive restructuring or mindful- ness, or whether positive effect relate to indirect effects like Additional Limitations and Strengths additional one-on-one attention for students from facilita- tors. These sorts of issues related to control conditions are A limitation of the research included in this overview relates already a criticism of depression prevention trials in wider to study settings and the generalizability of findings. Most youth settings. In addition, limited studies included meas- studies in included reviews were conducted in high income ures of positive mental health or wellbeing. The dual con- countries such as Australia, Canada, and the USA, limit- tinua model suggests that positive mental health and mental ing generalizability to other socio-economic and cultural ill-health operate on two spectrums rather than a single con- contexts. Understanding the impact of these differences tinuum (Iasiello et al., 2020). When studies only measure 1 3 Adolescent Research Review one of these spectrums they may fail to provide a complete programs were heterogenous across short and long-term picture of the impact of the intervention. Future studies outcomes. Importantly, the majority of reviews identified could benefit from adding measures of physical, social or for this overview combined findings from primary and sec- educational functioning. It would also be beneficial to draw ondary school settings and age groups, revealing a need upon multiple perspectives by including teachers or parents’ for well-designed systematic reviews that focus on these perspectives on the interventions, as the existing evidence groups separately. The identified gaps in the present review is based primarily on student self-report. Future research can inform future research including systematic reviews, as should aim to tease apart the effectiveness of active program well as recommendations and guidance that can be the basis components to better current understandings around which for improving youth mental health through school-based elements of programs are most effective. interventions. The current overview is also limited by the fact that it Supplementary Information The online version contains supplemen- did not include and compare findings from research in pri- tary material available at https://doi. or g/10. 1007/ s40894- 023- 00211-1 . mary school settings to those from secondary school set- tings. Results presented here provide an umbrella view of Acknowledgements The authors would like to acknowledge all con- tributors to the “Evidence Summary” that inspired this article (see research in secondary school settings, which can be help- reference: Zbukvic et al., 2020). ful for guiding future research and for informing guidance for decision-makers about different program and facilitator Authors’ Contributions IZ conceived of the study, participated in its design and coordination, interpretation of the data and drafted the types. It would be valuable for future reviews to critically manuscript; SM participated in the design and interpretation of the compare and contrast research from across primary, second- data and helped to draft the manuscript; SC participated in the interpre- ary, and indeed tertiary education settings, to fully address tation of the data and helped to draft the manuscript; RA participated developmental and contextual considerations that may influ- in the interpretation of the data and helped to draft the manuscript; VP participated in the interpretation of the data and helped to draft ence program effectiveness. the manuscript; LM participated in the interpretation of the data and A key strength of the present article was the attempt to helped to draft the manuscript; AB participated in the collection and answer a question that had not yet been addressed through interpretation of the data and helped to draft the manuscript; RP par- an overview regarding the effectiveness of school-based pro- ticipated in the interpretation of the data and helped to draft the manu- script; MT participated in the interpretation of the data and coordina- grams targeting anxiety and depression in adolescents. Find- tion helped to draft the manuscript. All authors read and approved the ings are highly relevant to the development of recommen- final manuscript. dations for decision-makers about secondary school-based mental health programs, where guidance is currently limited Funding Open Access funding enabled and organized by CAUL and its Member Institutions. and not always evidence-based. The wide reach of secondary schools globally represents an opportunity for international Declarations collaboration in research and guideline development to sup- port evidence-based decision making in this area. Conflict of interest The authors report no conflict of interest. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- Conclusion tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, Understanding the evidence for secondary school-based pro- provide a link to the Creative Commons licence, and indicate if changes grams targeting anxiety and depression—two of the most were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated common mental health challenges faced by young people— otherwise in a credit line to the material. If material is not included in represents an important area for early intervention. To date, the article's Creative Commons licence and your intended use is not systematic reviews have tended to combine findings from permitted by statutory regulation or exceeds the permitted use, you will research in primary and secondary school settings, produc- need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . ing unclear conclusions about program effectiveness for dif- ferent developmental periods. This overview presents the results of systematic reviews of research focused on second- ary school mental health programs. Overall, the findings are References encouraging and indicate that secondary school settings may be an appropriate context for effective mental health inter - Ahlen, J., Lenhard, F., & Ghaderi, A. (2015). Universal prevention for anxiety and depressive symptoms in children: A meta-anal- ventions targeting depression and anxiety outcomes. Results ysis of randomized and cluster-randomized trials. Journal of showed that targeted interventions for depression and anxi- Primary Prevention, 36(6), 387–403. https:// doi. org/ 10. 1 007/ ety are generally effective in the short term, though evidence s10935- 015- 0405-4 was less clear for long term outcomes; results for universal 1 3 Adolescent Research Review Bajgain, K. T., Amarbayan, M., Wittevrongel, K., McCabe, E., Naqvi, of Child Psychology & Psychiatry & Allied Disciplines, 56(7), S. F., Tang, K., Aghajafari, F., Zwicker, J. D., & Santana, M. 723–734. https:// doi. org/ 10. 1111/ jcpp. 12383 (2023). Patient-reported outcome measures used to improve Das, J. K., Salam, R. A., Lassi, Z. S., Khan, M. N., Mahmood, W., youth mental health services: a systematic review. Journal of Patel, V., & Bhutta, Z. A. (2016). Interventions for adolescent Patient-Reported Outcomes, 7(1), 14. https:// doi. org/ 10. 1186/ mental health: An overview of systematic reviews. Journal of s41687- 023- 00556-0 Adolescent Health, 59(4suppl), S49–S60. h t t p s : / / d o i . o r g / 1 0 . Ballard, M., & Montgomery, P. (2017). Risk of bias in overviews of 1016/j. jadoh ealth. 2016. 06. 020 reviews: A scoping review of methodological guidance and four- De Silva, S., Bailey, A. P., Parker, A. G., Montague, A. E., & Hetrick, item checklist. Research Synthesis Methods, 8(1), 92–108. https:// S. E. (2018). Open-access evidence database of controlled trials doi. org/ 10. 1002/ jrsm. 1229 and systematic reviews in youth mental health. Early Intervention Bastounis, A., Callaghan, P., Banerjee, A., & Michail, M. (2016). The in Psychiatry, 12(3), 474–477. https://doi. or g/10. 1111/ eip. 12423 effectiveness of the Penn Resiliency Programme (PRP) and its Dray, J., Bowman, J., Campbell, E., Freund, M., Wolfenden, L., Hod- adapted versions in reducing depression and anxiety and improv- der, R. K., McElwaine, K., Tremain, D., Bartlem, K., Bailey, J., ing explanatory style: A systematic review and meta-analysis. Small, T., Palazzi, K., Oldmeadow, C., & Wiggers, J. (2017). Journal of Adolescence, 52, 37–48. https://doi. or g/10. 1016/j. adole Systematic review of universal resilience-focused interventions scence. 2016. 07. 004 targeting child and adolescent mental health in the school set- Bell, I., Pot-Kolder, R. M., Wood, S. J., Nelson, B., Acevedo, N., Stain- ting. Journal of the American Academy of Child & Adolescent ton, A., & Allott, K. (2022). Digital technology for addressing Psychiatry, 56(10), 813–824. cognitive impairment in recent-onset psychosis: A perspective. Erford, B., Kress, V., Giguere, M., Cieri, D., & Erford, B. (2015). Schizophrenia Research: Cognition, 28, 100247. Meta-analysis: Counseling outcomes for youth with anxiety dis- Bennett, K., Rhodes, A. E., Duda, S., Cheung, A. H., Manassis, K., orders. Journal of Mental Health Counseling, 37(1), 63–94. Links, P., Mushquash, C., Braunberger, P., Newton, A. S., Kutcher, Feiss, R., Dolinger, S. B., Merritt, M., Reiche, E., Martin, K., Yanes, J. S., Bridge, J. A., Santos, R. G., Manion, I. G., McLennan, J. D., A., Thomas, C. M., & Pangelinan, M. (2019). A systematic review Bagnell, A., Lipman, E., Rice, M., & Szatmari, P. (2015). A and meta-analysis of school-based stress, anxiety, and depression youth suicide prevention plan for Canada: A systematic review of prevention programs for adolescents. Journal of Youth and Ado- reviews. Canadian Journal of Psychiatry, 60(6), 245–257. lescence, 48(9), 1668–1685. Bernecker, S. L., Coyne, A. E., Constantino, M. J., & Ravitz, P. (2017). Ferreira-Vorkapic, C., Feitoza, J. M., Marchioro, M., Simões, J., For whom does interpersonal psychotherapy work? A systematic Kozasa, E., & Telles, S. (2015). Are there benefits from teaching review. Clinical Psychology Review, 56, 82–93. yoga at schools? A systematic review of randomized control trials Bitsko, R. H., Holbrook, J. R., Ghandour, R. M., Blumberg, S. J., Vis- of yoga-based interventions. Evidence-Based Complementary & ser, S. N., Perou, R., & Walkup, J. T. (2018). Epidemiology and Alternative Medicine, 2015, 1. impact of health care provider-diagnosed anxiety and depression Fulambarkar, N., Seo, B., Testerman, A., Rees, M., Bausback, K., & among US children. Journal of Developmental and Behavioral Bunge, E. (2022). Review: Meta-analysis on mindfulness-based Pediatrics, 39(5), 395–403. https:// doi. org/ 10. 1097/ dbp. 00000 interventions for adolescents’ stress, depression, and anxiety in 00000 000571 school settings: A cautionary tale. Child and Adolescent Mental Bradshaw, M., Gericke, H., Coetzee, B. J., Stallard, P., Human, S., & Health. https:// doi. org/ 10. 1111/ camh. 12572 Loades, M. (2021). Universal school-based mental health pro- Galling, B., & Correll, C. (2018). Effectiveness of coordinated spe- grammes in low- and middle-income countries: A systematic cialty care for early psychosis [conference abstract]. Schizophrenia review and narrative synthesis. Preventive Medicine, 143, 106317. Bulletin, 44(suppl1), S108. https://doi. or g/10. 1093/ sc hbul/sb y015. https:// doi. org/ 10. 1016/j. ypmed. 2020. 106317 265 Brown, A., Rice, S. M., Rickwood, D. J., & Parker, A. G. (2016). Gates, M., Gates, A., Pieper, D., Fernandes, R. M., Tricco, A. C., Systematic review of barriers and facilitators to accessing and Moher, D., & Hartling, L. (2022). Reporting guideline for over- engaging with mental health care among at-risk young people. views of reviews of healthcare interventions: Development of the Asia-Pacific Psychiatry, 8(1), 3–22. https://doi. or g/10. 1111/ app y. PRIOR statement. BMJ, 378, 070849. https:// doi. org/ 10. 1136/ 12199bmj- 2022- 070849 Caldwell, D. M., Davies, S. R., Hetrick, S. E., Palmer, J. C., Caro, Gee, B., Reynolds, S., Carroll, B., Orchard, F., Clarke, T., Martin, P., López-López, J. A., & Welton, N. J. (2019). School-based D., & Pass, L. (2020). Practitioner Review: Effectiveness of indi- interventions to prevent anxiety and depression in children and cated school-based interventions for adolescent depression and young people: a systematic review and network meta-analysis. anxiety–a meta-analytic review. Journal of Child Psychology and The Lancet Psychiatry, 6(12), 1011–1020. https://doi. or g/10. 1016/ Psychiatry, 61(7), 739–756. s2215- 0366(19) 30403-1 Hetrick, S., Cox, G., & Merry, S. (2015). Where to go from here? Chahar, A., Gulati, R., Sharma, J., Likhar, N., & Dang, A. (2015). An exploratory meta-analysis of the most promising approaches Impact of community based therapy on children with men- to depression prevention programs for children and adolescents. tal health disorders: A health technology assessment. Value in International Journal of Environmental Research & Public Health, 18(3), 275. Health, 12(5), 4758–4795. https:// doi. org/ 10. 3390/ i jer p h1205 Cordier, R., Speyer, R., Mahoney, N., Arnesen, A., Mjelve, L. H., & 04758 Nyborg, G. (2021). Effects of interventions for social anxiety and Hetrick, S. E., Cox, G. R., Witt, K. G., Bir, J. J., & Merry, S. N. (2016). shyness in school-aged children: A systematic review and meta- Cognitive behavioural therapy (CBT), third-wave CBT and inter- analysis. PLoS ONE, 16(7), e0254117. https:// doi. org/ 10. 1371/ personal therapy (IPT) based interventions for preventing depres- journ al. pone. 02541 17 sion in children and adolescents. Cochrane Database of System- Cowen, S. L. (2014). A meta-analysis of school-based depression pre- atic Reviews. https://doi. or g/10. 1002/ 14651 858. CD003 380. pub4 vention programs for children and adolescents. Brigham Young Hetrick, S. E., O’Connor, D. A., Stavely, H., Hughes, F., Pennell, University. K., Killackey, E., & McGorry, P. D. (2018). Development of an Cristea, I. A., Mogoașe, C., David, D., & Cuijpers, P. (2015). Prac- implementation guide to facilitate the roll-out of early intervention services for psychosis. Early Intervention in Psychiatry, 12(6), titioner Review: Cognitive bias modification for mental health 1100–1111. https:// doi. org/ 10. 1111/ eip. 12420 PMID- 28177 191 problems in children and adolescents: a meta-analysis. Journal 1 3 Adolescent Research Review Hugh-Jones, S., Beckett, S., Tumelty, E., & Mallikarjun, P. (2021). Opinion in Psychiatry, 28(4), 336–342. https:// doi. org/ 10. 1097/ Indicated prevention interventions for anxiety in children and ado-YCO. 00000 00000 000170 lescents: A review and meta-analysis of school-based programs. Oldham-Cooper, R., & Loades, M. (2017). Disorder-specific versus European Child & Adolescent Psychiatry, 30(6), 849–860. generic cognitive-behavioral treatment of anxiety disorders in Hunt, H., Pollock, A., Campbell, P., Estcourt, L., & Brunton, G. children and young people: A systematic narrative review of evi- (2018). An introduction to overviews of reviews: Planning a rel- dence for the effectiveness of disorder-specific CBT compared evant research question and objective for an overview. Systematic with the disorder-generic treatment, Coping Cat. Journal of Child Reviews, 7(39), 1–9. and Adolescent Psychiatric Nursing, 30(1), 6–17. Iasiello, M., van Agteren, J., & Cochrane, E. M. (2020). Mental health Pascoe, M. C., Bailey, A. P., Craike, M., Carter, T., Patten, R. K., and/or mental illness: A scoping review of the evidence and impli- Stepto, N. K., & Parker, A. G. (2021). Single session and short- cations of the dual-continua model of mental health. Evidence term exercise for mental health promotion in tertiary students: A Base, 202(1), 1–45. https:// doi. org/ 10. 21307/ eb- 2020- 001 scoping review. Sports Medicine - Open, 7(1), 72. https://doi. or g/ James, A. C., James, G., Cowdrey, F. A., Soler, A., & Choke, A. (2015). 10. 1186/ s40798- 021- 00358-y Cognitive behavioural therapy for anxiety disorders in children Pearson, M., Chilton, R., Wyatt, K., Abraham, C., Ford, T., Woods, and adolescents. Cochrane Database of Systematic Reviews. H. B., & Anderson, R. (2015). Implementing health promotion https:// doi. org/ 10. 1002/ 14651 858. CD004 690. pub4 programmes in schools: A realist systematic review of research Kambara, K., & Kira, Y. (2021). School-based intervention for depres- and experience in the United Kingdom. Implementation Science, sive symptoms in upper secondary education: A focused review 10, 149. https:// doi. org/ 10. 1186/ s13012- 015- 0338-6 and meta-analysis of the effect of CBT-based prevention. Mental Pfeifer, J. H., & Allen, N. B. (2021). Puberty initiates cascading rela- Health & Prevention. https://doi. or g/10. 1016/j. mhp. 2021. 200206 tionships between neurodevelopmental, social, and internaliz- Karukivi, J., Herrala, O., Sateri, E., Tornivuori, A., Salantera, S., Aro- ing processes across adolescence. Biological Psychiatry, 89(2), maa, M., & Karukivi, M. (2021). The effectiveness of individual 99–108. https:// doi. org/ 10. 1016/j. biops ych. 2020. 09. 002 mental health interventions for depressive, anxiety and conduct Rocha, T.B.-M., Graeff-Martins, A. S., Kieling, C., & Rohde, L. A. disorder symptoms in school environment for adolescents aged (2015). Provision of mental healthcare for children and adoles- 12–18-A systematic review. Front Psychiatry, 12, 779933. https:// cents: A worldwide view. Current Opinion in Psychiatry, 28(4), doi. org/ 10. 3389/ fpsyt. 2021. 779933 330. Labelle, R., Pouliot, L., & Janelle, A. (2015). A systematic review Sancassiani, F., Pintus, E., Holte, A., Paulus, P., Moro, M. F., Cossu, and meta-analysis of cognitive behavioural treatments for suicidal G., Angermeyer, M. C., Carta, M. G., & Lindert, J. (2015). and self-harm behaviours in adolescents. Canadian Psychology/ Enhancing the emotional and social skills of the youth to promote psychologie Canadienne, 56(4), 368–378. https://d oi.o rg/1 0.1 037/ their wellbeing and positive development: A systematic review a0039 159 of universal school-based randomized controlled trials. Clinical Lee, R. S. C., Hermens, D. F., Scott, J., O’Dea, B., Glozier, N., Scott, E. Practice and Epidemiology in Mental Health, 11, 21–40. M., & Hickie, I. B. (2017). A transdiagnostic study of education, Shea, B. J., Reeves, B. C., Wells, G., Thuku, M., Hamel, C., Moran, employment, and training outcomes in young people with mental J., Moher, D., Tugwell, P., Welch, V., Kristjansson, E., & Henry, illness. Psychological Medicine, 47(12), 2061–2070. https:// doi. D. A. (2017). AMSTAR 2: A critical appraisal tool for systematic org/ 10. 1017/ S0033 29171 70004 84 reviews that include randomised or non-randomised studies of Levesque, R. J. R. (2016). Reviews in research on adolescence: Genres, healthcare interventions, or both. BMJ. https:// doi. org/ 10. 1136/ trends, and challenges. Adolescent Research Review, 1(1), 3–13. bmj. j4008 https:// doi. org/ 10. 1007/ s40894- 015- 0022-8 Solmi, M., Radua, J., Olivola, M., Croce, E., Soardo, L., Salazar, G., de Levinson, J., Kohl, K., Baltag, V., & Ross, D. A. (2019). Investigating Pablo, J., Shin, Il., Kirkbride, J. B., Jones, P., Kim, J. H., Kim, J. the effectiveness of school health services delivered by a health Y., Carvalho, A. F., Seeman, M. V., Correll, C. U., & Fusar-Poli, provider: A systematic review of systematic reviews. PLoS ONE, P. (2022). Age at onset of mental disorders worldwide large-scale 14(6), e0212603. https:// doi. org/ 10. 1371/ journ al. pone. 02126 03 meta-analysis of 192 epidemiological studies. Molecular Psychia- Ma, L., Zhang, Y., Huang, C., & Cui, Z. (2020). Resilience-oriented try, 27, 281–295. https:// doi. org/ 10. 1038/ s41380- 021- 01161-7 cognitive behavioral interventions for depressive symptoms in Stockings, E., Hall, W. D., Lynskey, M., Morley, K. I., Reavley, N., children and adolescents: A meta-analytic review. Journal of Strang, J., Patton, G., & Degenhardt, L. (2016). Substance use in Affective Disorders, 270, 150–164. young people 3: Prevention, early intervention, harm reduction, McGorry, P. D., & Mei, C. (2018). Early intervention in youth and treatment of substance use in young people. Lancet Psychia- mental health: Progress and future directions. Evidence Based try, 3(3), 280–296. Mental Health, 21(4), 182. https:// doi. or g/ 10. 1136/ ebmen Townshend, K., Jordan, Z., Stephenson, M., & Tsey, K. (2016). The tal- 2018- 300060 effectiveness of mindful parenting programs in promoting parents’ Meekums, B., Vicky Karkou, E., & Nelson, A. (2015). Dance move- and children’s wellbeing: A systematic review. JBI Database of ment therapy for depression. Cochrane Database of Systematic Systematic Reviews and Implementation Reports, 14(3), 139–178. Reviews. https:// doi. org/ 10. 1002/ 14651 858. CD009 895. pub2https:// doi. org/ 10. 11124/ JBISR IR- 2016- 2314 Moir, T. (2018). Why is implementation science important for interven- UNICEF. (2022). The Global Coalition for Youth Mental Well-being. tion design and evaluation within educational settings? Frontiers UNICEF. Retrieved March 2023 from https:// www. unicef. org/ in Education. https:// doi. org/ 10. 3389/ feduc. 2018. 00061partn ershi ps/ coali tion- youth- mental- wellb eing Nanthakumar, C. (2018). The benefits of yoga in children. Journal Weaver, L. L., & Darragh, A. R. (2015). Systematic review of yoga of Integrative Medicine, 16(1), 14–19. https:// doi. org/ 10. 1016/j. interventions for anxiety reduction among children and adoles- joim. 2017. 12. 008 cents. The American Journal of Occupational Therapy: Official National Academies of Sciences, E., and Medicine. (2019). The prom- Publication of the American Occupational Therapy Association. ise of adolescence realizing opportunity for all youth. National https:// doi. org/ 10. 5014/ ajot. 2015. 020115 Academies Press. https:// doi. org/ 10. 17226/ 25388 Werner-Seidler, A., Huckvale, K., Larsen, M. E., Calear, A. L., Maston, O’Dea, B., Calear, A. L., & Perry, Y. (2015). Is e-health the answer K., Johnston, L., Torok, M., O’Dea, B., Batterham, P. J., Susanne to gaps in adolescent mental health service provision? Current Schweizer, S., Skinner, R., Steinbeck, K., Ratcliffe, J., Oei, J. L., Patton, G., Wong, I., Beames, J., Wong, Q. J. J., Lingam, R., … 1 3 Adolescent Research Review Christensen, H. (2020). A trial protocol for the effectiveness of Younger, D. S. (2016). Epidemiology of childhood and adult mental digital interventions for preventing depression in adolescents: The illness. Neurologic Clinics, 34(4), 1023–1033. https://d oi.o rg/1 0. future proofing study. Trials, 21(1), 2. https:// doi. org/ 10. 1186/ 1016/j. ncl. 2016. 06. 010 s13063- 019- 3901-7 Zalsman, G., Hawton, K., Wasserman, D., van Heeringen, K., Arens- Werner-Seidler, A., Perry, Y., Calear, A. L., Newby, J. M., & Chris- man, E., Sarchiapone, M., Carli, V., Höschl, C., Barzilay, R., Bal- tensen, H. (2017). School-based depression and anxiety preven- azs, J., Purebl, G., Kahn, J. P., Sáiz, P. A., Lipsicas, C. B., Bobes, tion programs for young people: A systematic review and meta- J., Cozman, D., Hegerl, U., & Zohar, J. (2016). Suicide prevention analysis. Clinical Psychology Review, 51, 30–47. https:// doi. org/ strategies revisited: 10-year systematic review. The Lancet. Psy- 10. 1016/j. cpr. 2016. 10. 005 chiatry, 3(7), 646–659. https:// doi. org/ 10. 1016/ S2215- 0366(16) Werner-Seidler, A., Spanos, S., Calear, A. L., Perry, Y., Torok, M., 30030-X O’Dea, B., & Newby, J. M. (2021). School-based depression and Zbukvic, I., Cooke, S., & Anderson, R. (2020). Evidence summary: A anxiety prevention programs: An updated systematic review and review of school-based mental health programs. Orygen. https:// meta-analysis. Clinical Psychology Review, 89, 102079. https:// www. orygen. org. au/ Train ing/ Resou rces/ Depre ssion/ Evide nce- doi. org/ 10. 1016/j. cpr. 2021. 102079summar y/A-r eview-of- secon dar y-sc hool-based- ment al-healt h-p/ WHO. (2012, 2012). Adolescent mental health: mapping actions of orygen- schoo ls- evide nce- summa ry- pdf. aspx? ext= nongovernmental organizations and other international develop- ment organizations. World Health Organization. https://apps. who. Publisher's Note Springer Nature remains neutral with regard to int/ iris/ handle/ 10665/ 44875 jurisdictional claims in published maps and institutional affiliations. Yohannan, J., & Carlson, J. S. (2019). A systematic review of school- based interventions and their outcomes for youth exposed to trau- matic events. Psychology in the Schools, 56(3), 447–464. https:// doi. org/ 10. 1002/ pits. 22202 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Adolescent Research Review Springer Journals http://www.deepdyve.com/lp/springer-journals/evidence-for-targeted-and-universal-secondary-school-based-programs-TtiIrEHCsa

Loading next page...

References (61)

AC James (2015)
10.1002/14651858.CD004690.pub4
Cochrane Database of Systematic Reviews
B O’Dea (2015)
10.1097/YCO.0000000000000170
Current Opinion in Psychiatry, 28
Justina Yohannan, J. Carlson (2018)
A systematic review of school‐based interventions and their outcomes for youth exposed to traumatic events
Psychology in the Schools
Johanna Karukivi, Outi Herrala, Elina Säteri, Anna Tornivuori, S. Salanterä, M. Aromaa, K. Kronström, M. Karukivi (2021)
The Effectiveness of Individual Mental Health Interventions for Depressive, Anxiety and Conduct Disorder Symptoms in School Environment for Adolescents Aged 12–18—A Systematic Review
Frontiers in Psychiatry, 12
Brioney Gee, S. Reynolds, Ben Carroll, Faith Orchard, Tim Clarke, David Martin, Jon Wilson, Laura Pass (2020)
Practitioner Review: Effectiveness of indicated school-based interventions for adolescent depression and anxiety - a meta-analytic review.
Journal of child psychology and psychiatry, and allied disciplines
J. Ahlen, F. Lenhard, A. Ghaderi (2015)
Universal Prevention for Anxiety and Depressive Symptoms in Children: A Meta-analysis of Randomized and Cluster-Randomized Trials
The Journal of Primary Prevention, 36
M. Pearson, R. Chilton, K. Wyatt, C. Abraham, T. Ford, H. Woods, R. Anderson (2015)
Implementing health promotion programmes in schools: a realist systematic review of research and experience in the United Kingdom
Implementation Science : IS, 10
Robyn Feiss, Sarah Dolinger, Monaye Merritt, Elaine Reiche, Karley Martin, Julio Yanes, Chippewa Thomas, Melissa Pangelinan (2019)
A Systematic Review and Meta-Analysis of School-Based Stress, Anxiety, and Depression Prevention Programs for Adolescents
Journal of Youth and Adolescence, 48
A. Chahar, R. Gulati, J. Sharma, N. Likhar, A. Dang (2015)
Impact of Community Based Therapy on Children with mental health disorders: a health technology assessment
Value in Health, 18
F. Sancassiani, E. Pintus, A. Holte, P. Paulus, M. Moro, Giulia Cossu, M. Angermeyer, M. Carta, J. Lindert (2015)
Enhancing the Emotional and Social Skills of the Youth to Promote their Wellbeing and Positive Development: A Systematic Review of Universal School-based Randomized Controlled Trials
Clinical Practice and Epidemiology in Mental Health : CP & EMH, 11
P. McGorry, Cristina Mei (2018)
Early intervention in youth mental health: progress and future directions
Evidence-Based Mental Health, 21
Melissa Bradshaw, Hermine Gericke, B. Coetzee, P. Stallard, S. Human, M. Loades (2020)
Universal school-based mental health programmes in low- and middle-income countries: A systematic review and narrative synthesis.
Preventive medicine
S. Hetrick, Georgina Cox, S. Merry (2015)
Where to Go from Here? An Exploratory Meta-Analysis of the Most Promising Approaches to Depression Prevention Programs for Children and Adolescents
International Journal of Environmental Research and Public Health, 12
K Townshend (2016)
10.11124/JBISRIR-2016-2314
JBI Database of Systematic Reviews and Implementation Reports, 14
B Meekums (2015)
10.1002/14651858.CD009895.pub2
Cochrane Database of Systematic Reviews
Taryn Moir (2018)
Why Is Implementation Science Important for Intervention Design and Evaluation Within Educational Settings?
Frontiers in Education
S. Silva, Alan Bailey, A. Parker, Alice Montague, S. Hetrick (2018)
Open‐access evidence database of controlled trials and systematic reviews in youth mental health
Early Intervention in Psychiatry, 12
A. Bastounis, P. Callaghan, A. Banerjee, M. Michail (2016)
The effectiveness of the Penn Resiliency Programme (PRP) and its adapted versions in reducing depression and anxiety and improving explanatory style: A systematic review and meta-analysis.
Journal of adolescence, 52
E Stockings (2016)
10.1016/S2215-0366(16)00002-X
Lancet Psychiatry, 3
A. Werner-Seidler, K. Huckvale, M. Larsen, A. Calear, K. Maston, Lara Johnston, M. Torok, B. O’Dea, P. Batterham, S. Schweizer, S. Schweizer, S. Skinner, K. Steinbeck, J. Ratcliffe, J. Oei, G. Patton, Iana Wong, J. Beames, Quincy Wong, R. Lingam, K. Boydell, A. Salmon, N. Cockayne, A. Mackinnon, H. Christensen (2019)
A trial protocol for the effectiveness of digital interventions for preventing depression in adolescents: The Future Proofing Study
Trials, 21
C. Nanthakumar (2018)
The benefits of yoga in children.
Journal of integrative medicine, 16 1
K. Bennett, A. Rhodes, Stephanie Duda, A. Cheung, K. Manassis, P. Links, C. Mushquash, Peter Braunberger, A. Newton, S. Kutcher, J. Bridge, R. Santos, I. Manion, J. McLennan, A. Bagnell, E. Lipman, M. Rice, P. Szatmari (2015)
A Youth Suicide Prevention Plan for Canada: A Systematic Review of Reviews
The Canadian Journal of Psychiatry, 60
M. Solmi, J. Raduà, M. Olivola, E. Croce, L. Soardo, G. Pablo, Jae Shin, J. Kirkbride, Peter Jones, Jae Kim, Jong Kim, A. Carvalho, M. Seeman, C. Correll, P. Fusar-Poli (2021)
Age at onset of mental disorders worldwide: large-scale meta-analysis of 192 epidemiological studies
Molecular Psychiatry, 27
Madeleine Ballard, P. Montgomery (2017)
Risk of bias in overviews of reviews: a scoping review of methodological guidance and four‐item checklist
Research Synthesis Methods, 8
M. Gates, A. Gates, D. Pieper, R. Fernandes, A. Tricco, D. Moher, S. Brennan, Tianjing Li, M. Pollock, C. Lunny, Dino Sepúlveda, J. McKenzie, S. Scott, K. Robinson, K. Matthias, Konstantinos Bougioukas, P. Fusar-Poli, P. Whiting, S. Moss, L. Hartling (2022)
Reporting guideline for overviews of reviews of healthcare interventions: development of the PRIOR statement
The BMJ, 378
Kohei Kambara, Yugo Kira (2021)
School-based intervention for depressive symptoms in upper secondary education: A focused review and meta-analysis of the effect of CBT-based prevention
Mental Health & Prevention, 23
R. Levesque (2015)
Reviews in Research on Adolescence: Genres, Trends, and Challenges
Adolescent Research Review, 1
I. Cristea, Cristina Mogoașe, D. David, P. Cuijpers (2015)
Practitioner Review: Cognitive bias modification for mental health problems in children and adolescents: a meta-analysis.
Journal of child psychology and psychiatry, and allied disciplines, 56 7
B. Galling, Christoph Correll (2018)
O11.5. EFFECTIVENESS OF COORDINATED SPECIALTY CARE FOR EARLY PSYCHOSIS
Schizophrenia Bulletin, 44
B. Shea, Barnaby Reeves, George Wells, Micere Thuku, Candyce Hamel, Julian Moran, David Moher, P. Tugwell, Vivian Welch, Elizabeth Kristjansson, David Henry (2017)
AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both
The BMJ, 358
I Bell, RM Pot-Kolder, SJ Wood, B Nelson, N Acevedo, A Stainton, K Allott (2022)
Digital technology for addressing cognitive impairment in recent-onset psychosis: A perspective
Schizophrenia Research: Cognition, 28
SE Hetrick (2016)
10.1002/14651858.CD003380.pub4
Cochrane Database of Systematic Reviews
A. Werner-Seidler, S. Spanos, A. Calear, Y. Perry, M. Torok, B. O’Dea, H. Christensen, J. Newby (2021)
School-based depression and anxiety prevention programs: An updated systematic review and meta-analysis.
Clinical psychology review, 89
M. Iasiello, J. Agteren, E. Muir‐Cochrane (2020)
Mental Health and/or Mental Illness: A Scoping Review of the Evidence and Implications of the Dual-Continua Model of Mental Health
Evidence Base, 2020
Jai Das, Rehana Salam, Z. Lassi, Marium Khan, Wajeeha Mahmood, V. Patel, Z. Bhutta (2016)
Interventions for Adolescent Mental Health: An Overview of Systematic Reviews
The Journal of Adolescent Health, 59
C. Ferreira-Vorkapic, J. Feitoza, M. Marchioro, J. Simoes, E. Kozasa, S. Telles (2015)
Are There Benefits from Teaching Yoga at Schools? A Systematic Review of Randomized Control Trials of Yoga-Based Interventions
Evidence-based Complementary and Alternative Medicine : eCAM, 2015
Kalpana Bajgain, Mungunzul Amarbayan, Krystle Wittevrongel, E. McCabe, Syeda Naqvi, K. Tang, F. Aghajafari, Jennifer Zwicker, M. Santana (2023)
Patient-reported outcome measures used to improve youth mental health services: a systematic review
Journal of Patient-Reported Outcomes, 7
D. Caldwell, S. Davies, S. Hetrick, J. Palmer, Paola Caro, J. López-López, D. Gunnell, J. Kidger, James Thomas, C. French, E. Stockings, R. Campbell, N. Welton (2019)
School-based interventions to prevent anxiety and depression in children and young people: a systematic review and network meta-analysis
The Lancet. Psychiatry, 6
Adrienne Brown, S. Rice, D. Rickwood, A. Parker (2016)
Systematic review of barriers and facilitators to accessing and engaging with mental health care among at‐risk young people
Asia‐Pacific Psychiatry, 8
M. Pascoe, Alan Bailey, M. Craike, T. Carter, Rhiannon Patten, N. Stepto, A. Parker (2021)
Single Session and Short-Term Exercise for Mental Health Promotion in Tertiary Students: A Scoping Review
Sports Medicine - Open, 7
G Zalsman (2016)
10.1016/S2215-0366(16)30030-X
The Lancet. Psychiatry, 3
D. Younger (2016)
Epidemiology of Childhood and Adult Mental Illness.
Neurologic clinics, 34 4
H Hunt, A Pollock, P Campbell, L Estcourt, G Brunton (2018)
An introduction to overviews of reviews: Planning a relevant research question and objective for an overview
Systematic Reviews, 7
Youth Children, Families., E. Backes, R. Bonnie (2019)
The Promise of Adolescence
TB-M Rocha (2015)
10.1097/YCO.0000000000000169
Current Opinion in Psychiatry, 28
Lindy Weaver, A. Darragh (2015)
Systematic Review of Yoga Interventions for Anxiety Reduction Among Children and Adolescents.
The American journal of occupational therapy : official publication of the American Occupational Therapy Association, 69 6
Samantha Bernecker, A. Coyne, Michael Constantino, P. Ravitz (2017)
For whom does interpersonal psychotherapy work? A systematic review.
Clinical psychology review, 56
Bradley Erford, V. Kress, Monica Giguere, Domenic Cieri, Breann Erford (2015)
Meta-analysis: Counseling Outcomes for Youth with Anxiety Disorders
Journal of mental health counseling, 37
Julia Levinson, Kid Kohl, V. Baltag, D. Ross (2019)
Investigating the effectiveness of school health services delivered by a health provider: A systematic review of systematic reviews
PLoS ONE, 14
SE Hetrick (2018)
10.1111/eip.12420PMID-28177191
Early Intervention in Psychiatry, 12
R. Oldham-Cooper, M. Loades (2017)
Disorder‐specific versus generic cognitive‐behavioral treatment of anxiety disorders in children and young people: a systematic narrative review of evidence for the effectiveness of disorder‐specific CBT compared with the disorder‐generic treatment, Coping Cat
Journal of Child and Adolescent Psychiatric Nursing, 30
R. Bitsko, Joseph Holbrook, R. Ghandour, S. Blumberg, S. Visser, R. Perou, J. Walkup (2018)
Epidemiology and Impact of Health Care Provider–Diagnosed Anxiety and Depression Among US Children
Journal of Developmental & Behavioral Pediatrics, 39
J. Dray, J. Bowman, Elizabeth Campbell, M. Freund, L. Wolfenden, R. Hodder, Kathleen McElwaine, D. Tremain, K. Bartlem, J. Bailey, T. Small, K. Palazzi, C. Oldmeadow, J. Wiggers (2017)
Systematic Review of Universal Resilience-Focused Interventions Targeting Child and Adolescent Mental Health in the School Setting.
Journal of the American Academy of Child and Adolescent Psychiatry, 56 10
Jennifer Pfeifer, N. Allen (2021)
Puberty Initiates Cascading Relationships Between Neurodevelopmental, Social, and Internalizing Processes Across Adolescence
Biological Psychiatry, 89
A. Werner-Seidler, Y. Perry, A. Calear, J. Newby, H. Christensen (2017)
School-based depression and anxiety prevention programs for young people: A systematic review and meta-analysis.
Clinical psychology review, 51
Liang Ma, Yingnan Zhang, Cong Huang, Zeshi Cui (2020)
Resilience-oriented cognitive behavioral interventions for depressive symptoms in children and adolescents: A meta-analytic review.
Journal of affective disorders, 270
Nilija Fulambarkar, Brian Seo, A. Testerman, Mitchell Rees, Kimberly Bausback, E. Bunge (2022)
Review: Meta-analysis on mindfulness-based interventions for adolescents' stress, depression, and anxiety in school settings: a cautionary tale.
Child and adolescent mental health
R. Cordier, R. Speyer, Natasha Mahoney, A. Arnesen, L. Mjelve, Geir Nyborg (2021)
Effects of interventions for social anxiety and shyness in school-aged children: A systematic review and meta-analysis
PLoS ONE, 16
S. Hugh-Jones, Sophie Beckett, E. Tumelty, P. Mallikarjun (2020)
Indicated prevention interventions for anxiety in children and adolescents: a review and meta-analysis of school-based programs
European Child & Adolescent Psychiatry, 30
R. Labelle, L. Pouliot, A. Janelle (2015)
A systematic review and meta-analysis of cognitive behavioural treatments for suicidal and self-harm behaviours in adolescents
Canadian Psychology, 56
RSC Lee (2017)
10.1017/S0033291717000484
Psychological Medicine, 47

Publisher: Springer Journals
Copyright: Copyright © The Author(s) 2023
ISSN: 2363-8346
eISSN: 2363-8354
DOI: 10.1007/s40894-023-00211-1
Publisher site: See Article on Publisher Site

Abstract

While there are a number of systematic reviews on school-based mental health programs, there appears to be heterogeneity in their overall findings and conclusions, possibly due to the tendency to combine evidence from primary school and high school programs. To investigate the evidence for the effectiveness of targeted (for specific groups) and universal (for all students) mental health programs delivered in secondary schools, a systematic review of systematic reviews was conducted. A systematic search for reviews published from 2015 included outcomes for depression and anxiety—the most common mental health conditions—and quality appraisal of original studies in majority secondary school settings. A total of 395 references were screened and 14 systematic reviews were included. Of reviews that were assessed in full, most were excluded for not clearly separating studies in secondary school settings from primary school settings. Findings were generally positive but heterogenous in terms of quality and long-term outcomes. Interventions were mainly based on cognitive behavioural therapy (CBT), with results showing that targeted interventions are generally effective in the short-term for both anxiety and depression, while universal programs may be effective in some situations, typically in the short term and when programs were CBT-based; where reviews focused on universal resilience programs there were no significant effects. The overview shows a need for systematic reviews focused on secondary school settings, which consider contextual and individual factors that can influence the implementation and effectiveness of programs. Keywords Schools · Students · Adolescent · Humans · Mental Health Introduction in global research and public health campaigns (UNICEF, 2022), as well as the expansion of mental health services Depression and anxiety are amongst the most common men- focused on youth aged 12–25 years (Rocha et al., 2015). tal illnesses affecting adolescents and have been associated While the evidence for early intervention continues to accu- with social, school and family problems, poor physical mulate, there exists a paradox where adolescents experience health, and costs for individuals, families, and communi- some of the highest levels of mental health issues yet have ties (Bitsko et al., 2018). Early intervention and prevention the lowest levels of mental health service access of any age efforts for depression and anxiety aimed at adolescents have group (McGorry & Mei, 2018). This has driven an inter- become a high priority in many parts of the world, reflected est in strategies to support the mental health and wellbeing of young people outside of health systems. One increas- ingly common strategy is the use of school-based mental * Isabel Zbukvic health programs. There have now been several systematic isabel.zbukvic@orygen.org.au reviews published on school-based mental health programs, with mixed conclusions about effectiveness. Most of these Orygen, 35 Poplar Road, Parkville, VIC 3052, Australia 2 reviews, however, have combined research from across pri- Centre for Youth Mental Health, The University mary and secondary school settings, ignoring developmental of Melbourne, Victoria 3010, Australia 3 differences between these age groups as well as contextual Black Dog Institute, University of New South Wales, factors between primary and secondary education settings. Randwick, NSW 2031, Australia Vol.:(0123456789) 1 3 Adolescent Research Review The present overview of reviews seeks to address this gap by There has been much research to date on school-based summarising the range of existing evidence for high school mental health programs and several systematic reviews have based universal and selective mental health programs target- now been published on the topic. However, overall, there ing depression and anxiety. appears to be heterogeneity in their findings and conclu- Secondary school years capture a key developmental sions. This may be due to differences in methodological period for mental health, with global epidemiological data factors such as program type, delivery mode, implementing indicating that half of all lifetime mental disorders begin personnel (e.g., teachers, mental health clinicians, etc.), and by 18 years of age (Solmi et al., 2022). It is increasingly outcomes investigated. Mixed findings may also be because recognised that secondary schools are uniquely placed as many reviews do not separately consider the evidence from settings for effective intervention in youth mental health as primary school and high school programs. Given that the they offer a unique opportunity to reach a large number of prevalence of common mental illnesses increases substan- adolescents and overcome common barriers to help-seeking tially in adolescence as compared to childhood, reviews and and mental health service access such as stigma, physical meta-analyses that conflate the findings from primary and location and cost. Indeed, young people spend more time secondary school programs are ignoring critical develop- at school than in any other institution. The World Health mental considerations that may help to clarify whether and Organization (WHO) acknowledges that schools are well- why school-based programs are effective in different age positioned to support adolescent mental health by provid- groups (Younger, 2016). Overviews of systematic reviews— ing a setting for large-scale implementation of interventions, or, umbrella reviews—involve systematic review and syn- across a range of social contexts (WHO, 2012). In light of thesis of the evidence from multiple, existing systematic this, school-based mental health programs have proliferated reviews in a given area. They are a relatively recent approach in recent years. School-based mental health programs vary to evidence synthesis, and are considered particularly help- widely in terms of intervention type, delivery, the provi- ful for providing insight into the overall state of the evidence sion of ongoing support and involvement of young people. in situations where a number of systematic reviews have For the purposes of the present review, these programs are already been published evaluating interventions for a par- broadly classified into two types: mental health promotion ticular outcome (Hunt et al., 2018). While several systematic programs and mental illness prevention programs. Mental reviews on school based mental health programs targeting health promotion programs are those which focus on positive depression and anxiety have now been conducted, the pre- development and wellbeing, such as social inclusion, social sent review represents the first known umbrella review of and emotional skills, and problem-solving. Mental illness this area. prevention programs are those which focus on preventing the development of common mental disorders, such as anxiety and depression. This review will focus on prevention pro- Current Study grams for depression and anxiety, due to the prevalence of these conditions amongst adolescents and their far-reaching The existing evidence from systematic reviews presents and significant consequences. mixed findings for the effectiveness of school based mental School-based prevention programs can be classified into health programs addressing anxiety and depression, possibly two main approaches: universal approaches and targeted due to combining findings across primary and secondary approaches (Zbukvic et al., 2020). Universal approaches school populations. To address this, the current study aimed involve all students, regardless of their experience, identity, to provide an overview of systematic reviews of research or level of risk for developing a mental disorder. Targeted specifically in secondary schools. The present overview “selective” approaches are aimed at particular groups of identified the overall evidence for different program types, young people who are identified as having a higher risk of strengths and limitations of the evidence, and opportunities developing a mental disorder—for example, students who for future research. The research was guided by the overarch- live with parents experiencing mental ill-health, or students ing research question: What is the evidence for secondary with a neurodevelopmental condition. Targeted “indicated” school-based programs for improving depression and anxi- approaches are aimed at students who are already experi- ety outcomes in young people (aged 12–18 years)? encing identifiable symptoms below the threshold typically associated with a diagnosable mental disorder. The current overview will present the evidence from systematic reviews Methods focused on universal and/or targeted prevention programs in secondary schools, to help build knowledge about which This overview is presented using the preferred reporting program type may be most effective for addressing anxiety items for overviews of reviews (PRIOR) (Gates et al., 2022). and depression in adolescents. An internal protocol was developed to document the search 1 3 Adolescent Research Review strategy, eligibility criteria, data extraction and synthesis search strategy). The Evidence Finder’s anxiety and depres- procedures. The search strategy was pragmatically designed sion search retrievals were searched using the following to originally produce an “Evidence Summary” from key terms in title, abstract, keyword, and label fields: school findings (Zbukvic et al., 2020). The review was not reg- AND (systematic review OR (metaanalys OR meta-analys istered, but the published “Evidence Summary” provides OR meta analys)). Combined, the searching procedure pro- a record of the purpose and key methods used to produce vides coverage for records published from 1980 through to the review. For the present article, only systematic reviews 30 June 2022. were included; no supplementary primary studies or grey Titles and abstracts of retrieved records were then literature were searched or obtained, and all included articles searched in EndNote using the terms “anxiety” or “depres- were assessed for quality and risk of bias. Data collection sion”. Retrieved articles were screened for eligibility by forms and data extracted from the included studies can be abstract, using the eligibility criteria. Abstract and full text provided by the corresponding author on request. screening of articles was conducted by IZ, RA, SC, LM, VP and MT working independently using the eligibility criteria, Search Strategy with final eligibility for inclusion confirmed through discus- sion at team meetings. Where there was any hesitation or Literature was identified using the “Evidence Finder” disagreement about whether a review should be included, (https:// www. orygen. org. au/ Tr ain ing/ Evide nce- F inder), a this was resolved via discussion. publicly available living database and search engine jointly developed by Orygen and headspace (National Youth Mental Eligibility Criteria Health Foundation). The “Evidence Finder” is a comprehen- sive repository of peer-reviewed systematic reviews and con- Inclusion criteria were: systematic reviews of controlled trolled trials evaluating prevention and treatment approaches trials published in English focused on school-based inter- for common mental ill-health conditions with a peak onset ventions targeting prevention of, or reduction in, depression during adolescence/early adulthood (12–25 years). These and anxiety symptoms, with at least half the participant conditions include anxiety, depression, bipolar, psychosis, population (i.e., program recipients) comprising secondary eating disorders, substance-use, and suicide/self-harm. The school-aged young people (i.e., 12- to 18-year-olds). The co- systematic searching and screening methodology used to primary outcomes were anxiety and depression symptoms create and maintain the Evidence Finder is detailed in pub- and could be measured using any tool and at any time-point. lished studies (De Silva et al., 2018; Hetrick et al., 2018). School-based interventions were defined as programs deliv - Reproducible searches are run annually in MEDLINE, ered in secondary school settings, either in-curriculum or Embase and PsycINFO databases. Retrieved records are outside of usual classes (regardless of whether school sta ff or screened against pre-defined eligibility criteria, and included others delivered the intervention). No restriction was placed studies are coded within the database to support searching on intervention type (e.g., psychosocial, lifestyle interven- according to condition or disorder category, stage of illness, tions) or on comparison condition. Systematic reviews were intervention type, and study design. By July 2022, more than defined as using a systematic search strategy with or with- 480,000 articles had been retrieved and screened, and over out meta-analysis. Reviews were required to be published 5000 unique studies included in the Evidence Finder data- from 2015 onwards to focus on recent evidence and limit base, covering publication dates 1980 to 30 June 2021. The primary trial overlap, based on the assumption that literature “Evidence Finder” tool has been used as a searchable infor- included in recent reviews would cover much of the litera- mation source underpinning published systematic reviews, ture included in previously published reviews. scoping reviews and evidence mapping reviews (e.g., Bell Exclusion criteria was based on (i) evidence type (articles et al., 2022; Pascoe et al., 2021). were excluded if they were non-systematic reviews, primary For the present study, the team responsible for maintain- quantitative research studies, policy evaluations, editorials, ing the Evidence Finder (SC, AB) conducted searches in the conference papers, protocols, theses and book chapters), (ii) database for records with study design coded as “systematic participants (reviews were excluded if the mean age of par- review” AND “school” in the title, abstract, keyword or label ticipants was outside of 12–18 years or if less than half the fields. The original search was run in February 2019 and studies were identifiable as being conducted with secondary updated in July 2021 and again in August 2022, identifying school age young people age 12–18 years) (iii) interventions records published to 30 June 2021. To ensure coverage of (reviews were excluded if they described studies of inter- more recently published literature an additional search was ventions that did not target depression, anxiety or mental conducted in August 2022 of unscreened records retrieved health and/or if they less than half the studies were delivered for the Evidence Finder annual update for the period 1 July in a school-based setting) and (iv) outcomes (reviews were 2021 to 30 June 2022 (see Appendix for Evidence Finder excluded if primary outcomes of studies included alcohol 1 3 Adolescent Research Review or other drug use, psychosis, suicidality, physical health or trials (including any multiple counts) (N), and the number academic outcomes). Reviews were also excluded if they of systematic reviews (c) in the overview. Each “index” described interventions delivered in specialist secondary trial represents the first time a primary trial is included in schools. Where a review was later updated, the original an overview. CCA overlap scores are interpreted as slight review was excluded from results. (0–5), moderate (6–10), high (11–15) or very high (> 15). The CCA formula is as follows: Data Extraction and Synthesis N −r CCA = rc − r Six authors extracted data from included publications (IZ, LM, MT, RA, SC, VP) using a structured template spread- The AMSTAR 2 checklist was used to evaluate the qual- sheet. Extracted data included: author, year of publica- ity of included systematic reviews (Shea et al., 2017). The tion, title of publication, country/countries where primary checklist includes 16 items assessing methodological qual- research was conducted, aims, methodology, findings, con- ity. All SRs were assessed by one reviewer (LM, MT or VP) clusions, caveats and biases, and other comments. Results and a second reviewer (SM) duplicated the appraisal of 20% were collated into tables and synthesized into written sum- of SRs, with 93% agreement and only minor disagreements maries based on the approach to prevention (targeted, uni- that had no impact on cond fi ence grades. It is recommended versal), with results classified as statistically significant/ that review authors determine which of 16-items from the non-statistically significant and further by standardized AMSTAR 2 checklist are critical and non-critical for their effect size and 95% confidence interval (CI) or credible review and use these to grade included systematic reviews interval (CrI). Effect sizes were further described accord- (Shea et al., 2017). Based on this recommendation grades of ing to results of included articles. Results were classified confidence for each review were produced based on critical by time-point (immediately post-intervention or follow-up) flaws and non-critical weaknesses (see Table 3). and intervention type where relevant. A table providing an The checklist includes date of publication, to ensure overview of the characteristics of all included reviews was results are up to date. Systematic reviews published in the produced, including details of participants, interventions, last five years (e.g., 2017 onwards) were considered up to analysis and reported results (Table 1). A summary table was date. The final item is the match between the scope of the produced showing the proportion of the statistically signifi- overview and the primary trials within included systematic cant positive effects compared to null effects by intervention reviews. In this overview, only systematic reviews with a type and time-point (Table 2). The time-points for this sum- minimum of 50% of their primary trials fulfilling inclusion mary information were set to immediately post intervention, criteria for age/setting (secondary school) were included. 0–6 months post intervention, 6–12 post intervention and Systematic reviews with a higher proportion of primary 12 + months post intervention. studies fulfilling the inclusion criteria were considered a better match. Risk of Bias Risk of bias summaries across the primary trials included Results in each systematic review was extracted by LM, MT, SC and VP. An established four-item checklist for overviews Results of the Search of systematic reviews was used to assess potential bias both within and across included systematic reviews (Ballard & The original search (February 2019) retrieved 145 reviews, Montgomery, 2017). The four items include: (1) overlap, of which six were included (Bastounis, 2016; Caldwell et al., (2) rating of confidence from the AMSTAR 2 checklist, (3) 2019; Dray, 2017; Ferreira-Vorkapic, 2015; Sancassiani, date of publication, and (4) match between the scope of the 2015; Werner-Seidler, 2017). The updated searches (July included review and the overview itself. Each of these items 2021, August 2022) resulted in the retrieval and screening are described in detail below. of an additional 37 reviews and the inclusion of an additional It is also important to assess overlap in the use of the four articles. A supplementary search (August 2022) within primary studies across multiple systematic reviews, as high the unscreened search retrievals identified for the 2021–2022 overlap can contribute to biased results. The corrected cov- Evidence Finder update resulted in retrieval and screening of ered area (CCA) is a comprehensive and validated meas- an additional 231 reviews and inclusion of an additional five ure of overlap that has been used in a previous review of articles (Bradshaw et al., 2021; Fulambarkar et al., 2022; reviews of school-based interventions (Levinson et al., Kambara & Kira, 2021; Karukivi et al., 2021; Werner- 2019). The CCA is calculated based on three variables: the Seidler et al., 2021). Twenty-five reviews were assessed in number of “index” primary trials (r), the number of total full but not included because the age group did not meet 1 3 Adolescent Research Review 1 3 Table 1 Summary of characteristics and findings of included reviews First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity author ies (total in secondary evidence secondary preven- tion types parison frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) ticipants) or age (%) setting in > 50% Universal Targeted Universal Targeted studies) Brad- 12 studies 6 (50%) System- No pooled Universal Universal No inter- Unable to deter- N/A Unable to deter- N/A Considerable N/A shaw (15,893 atic analysis school- vention, mine pooled mine pooled heterogeneity (2021) partici- review based men- waitlist, efficacy efficacy in measures pants) tal health standard used to intervention cur- assess (CBT, riculum, depression positive active and anxiety psychol- control ogy, health pro- promotion gram, education, placebo social and control emotional wellbeing programs, life skills, other locally developed interven- tions) Bastou- 9 studies 6 (67%) Meta- No Universal Resilience Active Non-significant N/A Non-significant N/A Heterogeneity Sensitivity analysis nis (4,744 analy- programs control, effect effect for anxiety revealed age of (2016) partici- sis (CBT*, no inter- (immediate): (immediate): interven- participants as pants) IPT) vention, SMD = 0.13, MD = − 0.23, tion effects possible source of waitlist 95& CI 95% CI (I = 20%) heterogeneity [0–0.26] [− 1.09–0.62] and Non-significant depression effect on interven- depression: tion effects MD = 0.09, 95% (I = 65%) CI [− 0.61–0.80]; heterogeneity was lower after post- hoc sensitivity analysis, limiting analysis only to studies targeting the same age group, I2 = 36%, Tau2 = 0.25, df = 6, p = 0.15 Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Caldwell 137 studies 85 (62%) Meta- Yes—separate Universal Any universal No inter- Evidence for Evidence for exer- No evidence No evidence for Moderate het- Sensitivity analysis (2019) (56,620 analy- analyses for and or targeted vention, mindfulness cise from single (immediate) any intervention erogeneity conducted by partici- sis secondary targeted program waitlist, and relaxation trial (immediate; Weak evi- (6-12 m) removing studies pants) school age/ (CBT*, IPT, usual (immediate; (SMD − 0·47, dence for Evidence for CBT at high or unclear setting psycho- cur- SMD − 0·65, 95% CrI − 0·86 CBT-based from single risk of bias for supportive riculum, CrI − 1·14 to to − 0·09) interventions study (13-24 m; randomisation programs) attention − 0·19;) No evidence (6-12 m)— SMD − 0·50, and alloca- controls Weak evidence (6-12 m) third-wave 95% CrI − 0·96 tion conceal- for CBT Evidence for CBT CBT (SMD to − 0·05) ment; findings (immediate; from single study − 0·13, 95% unchanged SMD − 0·15, (13-24 m; SMD CrI − 0·27 CrI − 0·34 to − 0·26, 95% CrI to 0·01) and 0·04) − 0·52 to − 0·01) CBT + IPT No evidence (SMD − 0·10, (6–12 m) CrI − 0·26 to No evidence 0·05) (13–24 m) No evidence (13–24 m) Dray 57 studies 38 (67%) Meta- Yes—Meta- Universal Resilience Control Adolescent N/A Adolescent N/A Heterogeneity Not reported analysis programs (not trials: Non- trials: Non- for anxiety (2017) (41,521 analy- partici- sis possible for (CBT*, speci- significant significant interven- 11–18 year- Penn fied) or effects vs con- effects overall tion effects pants) olds Resiliency alternate trol (0-54 m): (0-54 m): (I = 84%) (‘adolescent Program, resil- SMD 0.02, SMD 0.05, and trials’) positive ience- 95% CrI –0.24 95% CrI depression psychology, focused to 0.20 or − 0.11 to 0.01 interven- mindful- inter- vs alternate or vs alternate tion effects ness) vention intervention: intervention: (I = 56%) SMD 0.12, CI SMD 0.08, CI -0.02 to 0.25 -0.08 to 0.24 Ferreira- 9 studies 7 (78%) Meta- No Universal Yoga-based Waitlist, Significant N/A Significant N/A Heterogeneity Not reported for anxiety Vork- (1,144 analy- programs physical effects effects apic partici- sis educa- (immediate): (immediate): interven- tion effects (2015) pants) tion, no SMD − 0·036, SMD − 0·42, inter- 95% CrI 95% CrI (I = 79%) − 0.76 to vention − 0.71 to and − 0.01 − 0.07 depression interven- tion effects (I = 17%) Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Feiss 42 studies 42 (100%) Meta- N/A Universal Any universal Waitlist, Significant effects (imme- Significant effects (immedi- High het- Not reported (2019) (7,310 analy- and or targeted usual diate): t(54) = − 3.72, ate): t(116) = − 3.120, erogeneity partici- sis targeted program care, p < 0.001, dexp = − 0.70, p < 0.01, dexp = − 0.62, for anxiety pants) (indi- (CBT*, attention 95%CIexp: − 0.94, − 0.46, 95%CIexp: − 0.81, − 0.43, interven- cated; stress support, dctrl = − 0.14, 95%CIc- dctrl = − 0.22, 95%CIc- tion effects selective inocula- inactive trl − 0.26, − 0.01 trl: − 0.34, − 0.10 I = 89.26%) by eth- tion*, medi- control, No significant follow-up effects (3-6 m): No significant follow-up effects and moder- nicity, tation, other active t(20) = − 0.72, p = 0.48, dexp = − 1.0, (3-8 m): t(78) = − 0.009, ate to high location, holistic control 95%CIexp: − 1.29 − 0.71, p = 0.99, dexp = − 0.56, among con- income) interven- (e.g. dctrl = − 0.77, 95%CIctrl: − 0.76, 0.48 95%CIexp − 0.81, − 0.31, trol effects tions wellness dctrl = − 0.56, 95%CIc- (I = 63.24%) course) trl: − 0.89, − 0.22 High heteroge- neity among depression interven- tion effects (I = 96.91%) and control effects (I = 95.07%) Fulam- 9 studies 9 (100%) Meta- N/A Universal Mindfulness- Usual cur- No significant N/A No significant N/A Studies were Not reported barkar (5,046 analy- based inter- ricula, effects overall effects overall conducted (2022) par- sis ventions waitlist, (0-11wks, (0-11wks, in varied ticipants) active CI 95% CI 95% settings, Note: control − .05,− .44) − .14,− .53) included a 1 study (e.g. range of par- (N = 145) yoga, ticipants e.g., examined relaxa- clinical and stress not tion, non-clinical anxiety/ health samples, and depres- cur- used various sion riculum, interventions outcomes sub- (Anxiety stance homogeneity use χ = 13.01, preven- p < .01 tion) Depression homogeneity χ = 29.91, p < .01) Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Gee 45 studies 36 (80%) Meta- No Targeted Manualised Passive N/A Significant effect N/A Significant Overall Excluding data (2020) (number analy- (indi- psycho- controls (immediate): effect (0-6 m): heterogeneity from three of partici- sis cated) logical (waitlist, SMD = − .49, 95% SMD = .34, 95% I = 81% depression- pants not programs usual CI − 0.79, − 0.19 CI − 0.48, − 0.21 focused cluster reported) (CBT or care, No significant No signifi- randomised trials, CBT- moni- follow-up effects cant effects the effect size for based* toring (0—> 12 m): (6—> 12 m): depression symp- includ- symp- SMD = -.45, 95% SMD = -.10, toms at postinter- ing Penn toms, CI − 1.34, 0.43 and 95% CI − 0.35, vention remained Resiliency psych- SMD = -.06, 95% 0.14 and in the small– Program, oeduca- CI − 0.26, 0.13 SMD = -.10, medium range IPT or IPT- tion), 95% CI − 0.28, (SMD = − .39, based, other self- 0.08 95% purely psy- help, CI − 0.54, − 0.25, chological indi- p < .001, k = 28) programs vidual support, active controls (not defined) Hugh- 20 studies 12 (60%) Meta- No Targeted Any universal No inter- N/A Significant effects N/A N/A Heterogeneity Studies with a low Jones (2076 analy- (indi- prevention vention, (0-12 m): I = 78% risk of contami- (2021) partici- sis cated) or early waitlist, g = − 0.28, 95% nation had an ES pants) intervention attention CI = − 0.50, − 0.05 of g = 0.03 (k = 5, program control and 12 m: 95% CI = − 0.14, (CBT*, play g = − 0.24, 95% 0.21) therapy, CI = − 0.48, 0.00 mindful- No significant ness, follow-up emotion effects (> 12 m): regulation, g = − 0.01, 95% working CI = − 0.38, 0.36 memory) Kambara 18 studies 18 (100%) Meta- N/A Universal Universal or No inter- N/A N/A No signifi- Significant effects Moderate Not reported cant effects (immediate): heterogeneity (2021) (3121 analy- and targeted vention, partici- sis targeted intervention waitlist, (immediate): g = − 0.56, in pairwise (g = − 0.04, 95%Cl comparison pants) (indi- programs usual cated) based on care, 95%Cl [− 0.78, − 0.34] for universal usual [− 0.27, 0.18] (I CBT* = 53.72%) cur- and selective riculum, (I = 52.12%) attention No significant follow-up effects Moderate Not reported control (3–12 + months) combined heterogeneity universal and selective approaches: (I = 51.58%) g = − 0.12, 95%Cl [− 0.30, 0.06] Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Karukivi 8 studies 8 (100%) System- N/A Targeted Targeted Placebo N/A Significant effects for N/A Significant effects Not reported N/A (2021) (425 atic (indi- interven- control, CBT (immediate, for CBT, partici- review cated) tions active 0–12 m) attention-bias- pants) (CBT*, control, No statistical analy- modification inter- no inter- ses performed for (immediate) personal vention, motivation-adap- 0–12 m) counselling, no tive-skills-trauma No significant attention- control resolution effects for bias modi- interpersonal fication, counselling motivation- (immediate, adaptive- 0-6 m) skills- No statistical anal- trauma yses performed resolution for motivation- adaptive- skills-trauma resolution Ma 38 studies 30 (79%) Meta- No Universal Resilience- Waitlist, N/A N/A Significant Significant effect Heterogeneity Not reported (2020) (24,135 analy- and oriented assess- effect (immediate): = 79.2% partici- sis targeted CBT inter- ment (immediate): g = 0.21, 95%Cl pants) (selec- ventions* only, g = 0.09, [ 0.06, 0.35] tive attention 95%Cl [ 0.02, And 0-6 m: g = 0.24, 95%Cl e.g. “at control, 0.16] No risk”; no inter- significant [ 0.10, 0.38] indi- vention, follow-up cated) placebo effects (0-6 m): g = 0.05, 95%Cl [- 0.03, 0.13] Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Sancas- 22 studies 13 (59%) System- No pooled Universal Emotional and Usual cur- No signifi- N/A No signifi- N/A Findings N/A siani (49,169 atic analysis social skills riculum, cant effects cant effects reported as (2015) partici- review programs no inter- (0–6 m) (0–6 m) very hetero- pants) (life skills vention, geneous for training, treat- age/grade of wellbeing ment as participants, programs) usual character- (not istics of speci- intervention fied), and control minimal groups, contact duration of (e.g. programs mailed and follow- reading up, outcomes materi- and tools als), physical educa- tion, other educa- tion or program Adolescent Research Review 1 3 Table 1 (continued) First No. of stud- No. of studies Level of Analysis of Type of Interven- Com- Results for ANXIETY outcomes (time- Results for DEPRESSION outcomes Heterogene- Sensitivity tion types parison author ies (total in secondary evidence secondary preven- frame, precision) (timeframe, precision) ity (statistical (year) no. of par- school setting school age/ tion (*included condition measure) in > 50% ticipants) or age (%) setting Universal Targeted Universal Targeted studies) Werner- 130 studies 90 (53%) Meta- Subgroup Univer- Psychological Waitlist, Effect sizes small Effect sizes small Moderate for Not reported Seidler (45,924 31% analy- analy- sal or or psychoe- no inter- (immediate; n = 72, g = 0.18, 95%CI: (immediate; n3 = 101, g = 0.21, both depres- (2021) partici- 10–14 years sis sis—no targeted ducational vention 0.12–0.26) 95%CI: 0.17–0.24)sion (I = 47, pants) (‘early ado- difference (indi- program e.g. (short-term; n = 29, g = 0.19, 95%CI: (short-term; n = 60, g = 0.17, 95%CI: 95%CI: lescent’) between cated; (CBT or usual 0.12–0.26) 0.13–0.22) 0.33–0.58) 45% size of selec- CBT- cur- (medium-term; n = 31, g = 0.23, 95% (medium-term; n = 48, g = 0.10, and anxiety 14–19 years effects by tive by based*, riculum, CI: 0.10–0.35) 95%CI: 0.06–0.13) (I = 46, (‘older ado- age “risk” IPT, mind- active (long-term; n = 8, g = 0.11, 95%CI: (long-term; n = 20, g = 0.10, 95%CI: 95%CI: lescents’) e.g. fulness, control 0.03–0.18) 0.05–0.15) 0.29–0.59) parental personality- (not No significant No significant differ - Significantly Significantly larger psycho- focus, other speci- differences at ences at immedi- smaller effect size com- pathol- therapeutic fied) immediate, ate, short, medium effect size pared to univer- ogy, approaches) short, medium or long-term compared sal (immediate; person- or long-term follow up between to targeted n = 43 g = 0.29, ality, follow up universal and tar- (immedi- 95%CI: behav- between geted programs ate; n = 58, 0.22–0.37) ioural universal g = 0.17, No significant dif- prob- and targeted 95%CI: ferences at short, lems) programs 0.13–0.21) medium or long- No significant term follow up differences between univer- at short, sal and. targeted programs medium or long-term follow up between universal and. targeted programs CBT cognitive behavioural therapy, IPT interpersonal therapy indicates majority (> 50%) of programs. Where specific interventions are included in results columns, this reflects where significant findings were only reported for those interventions Adolescent Research Review Table 2 Summary of review findings by outcome, approach, and effectiveness over time Anxiety Depression Universal Targeted Universal Targeted n % CCA n % CCA n % CCA n % CCA Immediate positive effects 6 66.7 (60.0) 12.7 5 100.0 (100.0) 12.9 5 45.5 (45.5) 9.3 6 100.0 (85.7) 12.0 Immediate no effect 4 44.4 (40.0) 4.3 0 0.0 (0.0) NA 6 54.5 (54.5) 4.6 1 16.7 (14.3) NA 0–6 months positive effect 2 22.2 (28.6) 9.8 2 40.0 (40.0) 0.0 2 18.2 (22.2) 37.4 3 50.0 (42.9) 10.0 0–6 months no effect 5 55.6 (71.4) 4.2 3 60.0 (60.0) 11.8 7 63.6 (77.8) 5.8 4 66.7 (57.1) 5.5 6–12 months positive effect 2 22.2 (50.0) 9.8 2 40.0 (50.0) 0.0 2 18.2 (40.0) 37.4 2 33.3 (50.0) 0.0 6–12 months no effect 2 22.2 (50.0) 13.7 2 40.0 (50.0) 13.8 3 27.3 (60.0) 2.1 2 33.3 (50.0) 4.1 > 12 months positive effect 1 11.1 (25.0) NA 2 40.0 (66.7) 37.4 1 9.1 (20.0) NA 2 33.3 (66.7) 37.4 > 12 months no effect 3 33.3 (75.0) 9.4 1 20.0 (33.3) NA 4 36.4 (80.0) 8.4 1 16.7 (33.3) NA Number of SRs on topic 9 5 11 6 All results positive effect 40.9 64.2 31.9 61.3 All results no effect 59.1 35.8 68.1 38.7 Percentages show the proportion of reviews reporting the relevant effect out of the total number of reviews on that topic. Percentages in paren- theses are adjusted for the number of relevant effect sizes at that time point, as some reviews reported multiple effects for a single time point when comparing different types of interventions. Positive effects include any statistically significant positive effects regardless of effect size. n the number of relevant effects reported by reviews for each effect type. CCA Corrected Covered Area for the specific effect calculated based on the number of reviews and associated included primary studies reporting that effect. NA Not Applicable and is shown where it was not possible to calculate the CCA due to only one or less reviews reporting the relevant effect Adjusted percentage based on overall number of effects reported across all time points Table 3 AMSTAR 3 checklist quality appraisal of included studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Overall score Bastounis (2016) + + + + + + + + + − + + + + − − Moderate quality Bradshaw (2021) + + + + + + + + + − NMA NMA + + NMA − Moderate quality Caldwell (2019) + ± + ± + + + + + − + + + + + + High quality Dray (2017) + + + ± + + + + + − + + + + + + High quality Feiss (2019) + + + ± + + + + + − + − + + − + Moderate quality Ferreira-Vorkapic (2015) + − + ± + + + ± ± − NMA NMA + + NMA + Moderate quality Fulambarkar (2022) + − − ± + − + + + − + − − − − − Critically-low quality Gee (2020) + + + ± + + + ± ± − + − + + + + Moderate quality Hugh-Jones (2021) + + + + + + + + + − + + + + + + High quality Kambara (2021) + − + + + + + + + − + ± + + + + Moderate quality Karukivi (2021) + − + ± + − ± ± + − NMA NMA + − NMA + Critically-low quality Ma (2020) + − − ± − + + + + − + − + + + + Low quality Sancassiani (2015) + − + − − − − + - − NMA NMA + + NMA + Critically-low quality Werner-Seidler (2021) + + + + + + + + + − + + + + + + Moderate quality High Quality = Zero or one non-critical weakness. Moderate quality = More than one non-critical weakness. Low quality = One critical flaw with or without non-critical weaknesses. Critically low = More than one critical flaw with or without non-critical weaknesses. NMA = Not a meta- analysis. Item 1 in the checklist refers to: Did the research questions and inclusion criteria for the review include the components of PICO? criteria (Cordier et al., 2021), the settings of included studies include depression or anxiety (Bennett, 2015; Das, 2016; were not majority secondary school (Bennett, 2015; Cristea, Labelle, 2015; Lee et al., 2017; Meekums, 2015; O'Dea, 2015; Erford, 2015; James, 2015; Labelle, 2015; Oldham- 2015; Weaver, 2015; Zalsman, 2016), studies were not Cooper, 2017; Townshend, 2016) or could not be confirmed quality-appraised (Bernecker, 2017; Nanthakumar, 2018), as majority secondary school setting (Ahlen, 2015; Hetrick, or the record was an abstract not a review paper article (Cha- 2015; Hetrick, 2016; Stockings, 2016), outcomes did not har, 2015; Cowen, 2014; Galling, 2018). One review was 1 3 Adolescent Research Review updated in the time that searches were conducted so the orig- et al., 2021). Table 1 provides an overview of all included inal review has been excluded from results (Werner-Seidler, reviews with key findings for depression and anxiety out- 2017). Figure 1 provides an overview of the exclusion and comes, number of included studies and participants, pro- inclusion of reviews. gram and intervention types, and comparison conditions. The majority of studies focused on CBT-based interven- Study Characteristics tions, but reviews included research on a range of inter- ventions including health promotion, life and social skills, In total, fourteen systematic reviews were included that resilience programs, yoga-based programs, interpersonal considered findings for school-based mental health pro- therapy-based programs, emotion regulation and other grams with outcomes for anxiety and/or depression psychological and physical health-based interventions. (Bastounis, 2016; Bradshaw et al., 2021; Caldwell et al., Most significant findings related to reviews of majority 2019; Dray, 2017; Feiss et al., 2019; Ferreira-Vorkapic, CBT-based programs. Where reviews focused on univer- 2015; Fulambarkar et al., 2022; Gee et al., 2020; Hugh- sal resilience programs, there were no significant effects. Jones et al., 2021; Kambara & Kira, 2021; Karukivi et al., All reviews included studies with a range of comparison 2021; Ma et al., 2020; Sancassiani, 2015; Werner-Seidler Fig. 1 Flow diagram for inclu- Records identified from systematic Records removed before sion and exclusion of reviews reviews search screening Evidence Finder databse (n=183) Duplicates (n=19) Evidence Finder unscreened search retrievals (n=231) Records screened From systematic reviews search (n=395) Records sought for retrieval From systematic reviews search (n=38) Records assessed for retrieval Records excluded From systematic reviews search (n= From systematic reviews search: 38) Outcomes not depression or anxiety (n=8) Total included Setting not majority secondary school (n=7) Systematic reviews search (n=14) Setting could not be confirmed as majority secondary school (n=4) Not a review paper (n=3) Papers included in review not quality appraised (n=2) Age group did not meet criteria (n=1) 1 3 Adolescent Research Review conditions including no intervention, waitlist, attention reported no effect (Bastounis, 2016; Sancassiani, 2015). Lik - control, and many compared to usual curriculum. wise for anxiety, there was some evidence for effectiveness of universal programs short term (Ferreira-Vorkapic, 2015) Description of Review Findings while others found no effect (Bastounis, 2016; Sancassiani, 2015). There were reasonably consistent positive findings for Although the present overview only included systematic targeted programs from reviews that combined results from reviews of research majority conducted in secondary school primary and secondary settings in the short-term, however settings, most reviews combined findings from across sec- these were not maintained beyond 6–12 months (Gee et al., ondary and primary school participants or settings. Out of 2020; Hugh-Jones et al., 2021; Ma et al., 2020). fourteen reviews, four focused on studies specifically in sec- Overall, there was more evidence available for univer- ondary school settings (Feiss et al., 2019; Fulambarkar et al., sal than targeted interventions, and only a small number of 2022; Kambara & Kira, 2021; Karukivi et al., 2021). Of reviews directly compared these approaches. Table 2 pro- those four, two reviews examined the effectiveness of uni- vides a summary of the overarching findings across reviews versal programs and found no significant effects overall for and shows that evidence was strongest for the efficacy of tar - either anxiety or depression outcomes (Fulambarkar et al., geted interventions for both depression and anxiety, particu- 2022; Kambara & Kira, 2021) while two reviews that exam- larly directly after the intervention, with greater heterogene- ined targeted indicated programs found significant effects for ity in longer term effects. In contrast, universal programs anxiety and depression outcomes for CBT-based programs were found to be effective in less than half of the reviews up to 12 months (Kambara & Kira, 2021; Karukivi et al., on anxiety, and just over one-third for depression, with less 2021) and one review that combined findings from univer - support again for the efficacy of such programs long term. sal and targeted programs found significant effects in the Across reviews, facilitators included health workers and short- but not long-term (Feiss et al., 2019). Three reviews doctors, researchers, school counsellors, teachers, psycholo- conducted separate (Caldwell et al., 2019; Dray, 2017) or gists, graduate students, and other trained instructors. There sub-analyses (Werner-Seidler et al., 2021) with adolescent was no evidence for different outcomes between mental age groups. From these findings, there was weak evidence health professionals or teachers in two reviews (Bastounis, for CBT-based programs as a universal intervention for anxi- 2016; Caldwell et al., 2019), while another review found that ety and depression up to 12 months, and no evidence for interventions were more effective, especially at longer term effectiveness beyond a year (Caldwell et al., 2019), while follow ups, when delivered by mental health professionals another review found no significant evidence for universal (Ma et al., 2020). Another review found that indicated inter- programs in secondary settings (Dray, 2017). These reviews ventions delivered by teachers had no impact on depression also produced some limited evidence for CBT-based pro- or anxiety symptoms (Gee et al., 2020). The original and grams as a targeted intervention for both anxiety and depres- updated review published by one group suggested a slight sion, but where results were combined for universal and tar- preference for external personnel over school staff, though geted interventions in a review focused on secondary school, the earlier review found no significant difference between effects were limited to the first 3 months post-intervention groups for anxiety outcomes, and across both reviews’ find- (Feiss et al., 2019). No reviews compared the effectiveness ings were not maintained at longer-term follow-up (Wer- of targeted or universal approaches in primary vs secondary ner-Seidler et al., 2021; Werner-Seidler, 2017). Digital and school settings, however, one review conducted subgroup face-to-face approaches also appear to be similar in terms analysis and found no difference between effect sizes by age of effectiveness, although there is less evidence currently (Werner-Seidler et al., 2021). That review found small effect available for digital interventions (Caldwell et al., 2019; Gee sizes for universal and targeted programs overall across mul- et al., 2020; Werner-Seidler et al., 2021). Most studies in tiple time-points beyond 12 months after intervention. When included reviews were conducted English-speaking coun- compared using subgroup analyses, there were no signifi- tries including Australia, Canada, and the USA. cant differences between universal and targeted programs for anxiety, while for depression there were significantly smaller Quality Appraisal of Included Systematic Reviews effect sizes for universal programs compared to targeted in the short term, though these were not maintained at follow- The CCA was found to be 5.98%, which represents a moder- up (Werner-Seidler et al., 2021). ate overlap between the 14 systematic reviews covering 319 For reviews that combined results from research with studies. Detailed explanations and calculations for the CCA primary and secondary school students without separate can be found in the Appendix. Table 3 presents the checklist analyses, there was mixed evidence for universal programs. for quality appraisal of included reviews. The majority of For depression outcomes, some reported significant effects reviews were of moderate to high quality, although there was (Ferreira-Vorkapic, 2015; Ma et al., 2020) while others one low and three critically-low quality reviews. Of note, 1 3 Adolescent Research Review out of the four reviews that focused on secondary school reviews of research on school-based universal or targeted programs, two were of critically low quality and two moder- interventions for depression or anxiety, with findings from ate quality. A detailed overview of the AMSTAR 2 review generally high-quality reviews suggesting that programs can rating checklist and quality appraisal of primary studies can be effective short-term but benefits may diminish over time. be found in the Appendix. Table 4 shows the target outcome Overall, the results of the present review show that tar- of each included systematic review along with the results geted school-based interventions are generally effective at for other 3 items from the checklist for assessing risk of reducing depression and anxiety symptoms in young peo- bias in overviews of reviews. All but three of the systematic ple, especially at the time of intervention, while universal reviews were considered up to date by AMSTAR criteria of programs show greater heterogeneity in results. Regardless published in the last five years. The proportion of included of intervention or program type, evidence for the long-term relevant primary studies targeting secondary school age efficacy of school-based interventions for depression and young people ranged from 50 to 100%, with four reviews anxiety is still mixed and highly variable. It is not clear including 100% relevant studies. which programs may be most suitable for secondary school students, as most reviews combined findings from across age groups and settings. The combined results provide some Discussion basis for integrating universal or targeted interventions for anxiety and depression into school health and wellbeing pro- There have now been several systematic reviews of school- grams. However, the heterogeneity in findings and quality based mental health programs aimed at anxiety and depres- across the reviews limits strong conclusions. In particular, sion published without clear conclusions for effectiveness, limitations in the participant cohorts, study designs, and pro- possibly due to combining findings from primary and sec- gram implementation strategies leave a number of questions ondary school settings without considering the develop- and opportunities for future investigation. mental differences between age groups, or the heightened period of vulnerability for the emergence of anxiety and Evidence for Different Program and Facilitator Types depression symptoms during secondary school years. This umbrella review sought to address this issue by undertaking The present overview provides a summary of the different a developmentally focused examination of existing evidence program and facilitator types that have been evaluated in for programs delivered in secondary schools specifically. school-based mental health programs. Consistent with pre- The current umbrella review included fourteen systematic vious research, results suggest that targeted programs may Table 4 Risk of Bias assessment for included systematic reviews First Author Target outcome AMSTAR 2 Rating (item 2) Publication % of included studies rel- b c date (item 3) evant by age/setting (item 4) (%) Bastounis Depression Moderate quality 2016 67 Bradshaw Anxiety & depression Moderate quality 2021 50 Caldwell Anxiety & depression High quality 2019 62 Dray Resilience High quality 2017 67 Feiss Anxiety & depression Moderate quality 2019 100 Ferreira-Vorkapic Anxiety & depression Moderate quality 2015 78 Fulambarkar Anxiety, depression & stress Critically low quality 2022 100 Gee Anxiety & depression Moderate quality 2020 80 Kambara Depression Moderate quality 2021 100 Karukivi Anxiety, depression & conduct disorder Critically low quality 2021 100 Hugh-Jones Anxiety High quality 2021 60 Ma Depression Low quality 2020 79 Sancassiani Wellbeing, positive development, healthy Critically low quality 2015 59 lifestyle & academic performance Werner-Seidler Anxiety & depression Moderate quality 2021 76 See appendix for a detailed overview of the AMSTAR 2 rating for each included systematic review All reviews published since 2017 were considered up to date Proportion of studies within the systematic review that clearly met the current study inclusion criteria 1 3 Adolescent Research Review be more effective than universal programs (Werner-Seidler efficiency outcomes, which can build relevant knowledge to et al., 2021). Although it appears that targeted approaches support decision-makers (Moir, 2018). may be more effective than universal, it is essential to acknowledge that the direct comparison of different pro- Developmental Considerations gram types is threatened by confounding factors, particularly symptom levels at baseline. Universal programs include all The present overview aimed to understand the evidence individuals in a cohort, regardless of mental health needs, for secondary school mental health programs by including whereas targeted interventions focus on those who have systematic reviews where it was possible to identify that at greater mental health support needs and therefore have least 50% of primary studies were conducted in a second- greater potential for improvement relative to universal pro- ary school setting or with secondary school age adolescents. grams. Such design characteristics make it difficult to effec- All but one included > 60% studies with secondary school tively compare universal and targeted programs, and may age children, suggesting the majority of primary studies in lead to less positive conclusions about the efficacy of uni- included reviews aligned with the overview scope. How- versal interventions than are warranted (Feiss et al., 2019). ever, the mixture of findings from primary and secondary Overall, while findings from the present overview helps to school studies in most included reviews means that results paint a picture of the evidence for school-based programs should still be interpreted with some caution. While more targeting secondary school students, they also help to show recent reviews (e.g., Werner-Seidler et al., 2017, 2021) have where there are current gaps in the literature and opportuni- addressed this to some extent using sub-group analyses, ties for future research. there is a clear need for systematic reviews that focus on Programs studied in the included reviews were primar- each age group and setting. The present overview offers a ily cognitive behavioural therapy-based. While these pro- step in this direction, presenting the evidence from system- grams appear to be the most effective targeting depression atic reviews of majority secondary school-based research. and anxiety, such conclusions can only be made tentatively The developmental differences between primary school as more research on non-CBT programs is needed. There age (5–11 years) and secondary school age (12–18 years) are was mixed evidence for the impact of facilitators across significant, particularly in relation to mental health preven- the reviews, with most delivered by mental health profes- tion and treatment needs. These age groups show differences sionals or teachers. Taken together, the reviews suggest in cognitive abilities, neurobiology, relationships with peers that interventions may be more effective when delivered by and family; there are also significant differences that come mental health professionals rather than teachers, especially following puberty in terms of romantic relationships, explo- for indicated interventions. Results also suggested similar ration of sexuality and gender, and bodily changes, as well effectiveness for programs delivered in person compared to as new social pressures related to academic performance and online, however the research in this area was limited. At working towards independent living, all of which can impact least one randomized controlled trial is currently underway and be impacted by mental health and wellbeing (National focused on understanding the effectiveness of online app- Academies of Sciences, 2019; Pfeifer & Allen, 2021). These based interventions for mental health delivered in secondary developmental differences have been proposed to explain school settings, which will add to the evidence in this area the rise in rates of mental ill-health between childhood and (Werner-Seidler et al., 2020). From the present findings, it adolescence. It is perhaps not surprising that targeted pro- is not clear under what conditions specific facilitators and grams show stronger evidence for effectiveness in this age delivery methods can have the greatest impact. group compared to findings for universal programs, where Future research designs should consider the need to targeted programs can both address the specific needs of understand how, why, and when secondary school programs adolescents (e.g., peer pressure, academic stress) as well can be most effective. The importance of separating research as being more likely to produce positive impacts on symp- in primary and secondary school settings is relevant here toms if starting from a higher baseline. Results of the pre- too, with major differences in learning environment, pro - sent overview showed that some reviews combining results cesses and structures that can all impact implementation, for primary and secondary school settings reported positive which can in turn influence the reach and impact of pro- results for universal programs, whereas reviews focused grams (Pearson et al., 2015). Although not the focus of the specifically on secondary school did not. It is possible that current review, an understanding of the evidence for cost- universal programs may be more suitable and effective for effectiveness will also help with decision-making about primary school students, which may be confounding results future investment in universal and/or targeted school-based for reviews that combined age groups and settings. This mental health programs. The e fi ld of implementation science may be due to depression and anxiety symptoms being offers frameworks to help capture factors that may influ- more amenable to universal intervention earlier in life, or ence implementation in school settings, as well as cost and it may be related to differences between settings making 1 3 Adolescent Research Review implementation more effective. Future research could help is crucial, as one review found that found that programs to disentangle these questions. The mixed and weaker effects delivered in lower socioeconomic status secondary school for long-term effects for school-based programs overall may settings were less effective than in high or mixed socioeco- also be interpreted through a developmental lens, consider- nomic schools (Caldwell et al., 2019). Similarly, assessing ing the rapid and significant changes that occur for adoles- the effectiveness and acceptability of school-based interven- cent over periods of one year or more. It may be the case tions for depression and anxiety for those from culturally that targeted programs benefit from being delivered regularly and linguistically diverse and LGBTQI + communities, is across secondary school years, however, further research is vital. Such groups often face unique stressors and barriers needed to support this approach. to help seeking that can exacerbate mental health problems, and may require tailored or adapted interventions to best Limitations of Overview Review Methodology support their needs (Brown et al., 2016). Future research and systematic reviews must consider the contextual and Overviews of systematic reviews can provide a comprehen- individual factors that can influence the implementation and sive summary of the existing evidence on a topic, but also effectiveness of school-based programs. Those responsible come with limitations (Ballard & Montgomery, 2017). Over- for program design, implementation, and evaluation should views are unlikely to include the most recent primary studies aim to capture inclusive data about student identity and due to the time lag between study publication and inclusion background, and to assess programs delivered in a range of in systematic reviews. Therefore, it is possible that more settings. Where the evidence is limited, designers and facili- recent studies have addressed the identified research gaps in tators should adapt school-based mental health programs this study. Additionally, scope mismatch, or when the scope according to relevant guidelines and models to ensure cultur- of included systematic reviews is not directly aligned with ally responsive practice (Yohannan & Carlson, 2019). the overview, can also pose a problem, as conclusions may Another limitation of included research relates to study be based on studies that do not fully address the scope of design and outcome measures. While long-term outcomes the overview. Finally, overlap of primary studies between were mixed overall, interpretation is somewhat limited by included systematic reviews can bias results and conclu- long-term impacts not being consistently measured by the sions of overviews. The current study used a CCA analysis, same tools. The development of meaningful, appropriate demonstrating a moderate level of overlap in primary stud- youth mental health measures and outcomes is an area of ies. Critically, overviews can also only provide conclusions research focus that will help with future school-based men- that are accurate when the included reviews are of a high tal health program evaluation and research (Thapa Bajgain standard. Four of the reviews in this overview were critically et al., 2023). Consensus on measures and study designs will low- or low-quality confidence based on the AMSTAR 2 help to promote consistency across the field and enhance checklist, which included two of the four reviews focused on opportunities to identify the key mechanisms that could be secondary school research, with the rest of either moderate leveraged to improve program delivery and efficacy. Con - (n = 7) or high confidence (n = 3). Given the stringency of the sistency in the measures used to assess mental health out- checklist, this suggests that overall quality of the included comes would also support comparison between approach reviews is relatively good (Levinson et al., 2019), however and program types. Without such information, policy mak- the poor quality found for secondary school research casts ers, researchers, and leaders in school health are limited by some doubt on the conclusions that can be drawn from those the evidence-base in terms of recommendations for imple- reviews. Results of the present overview must be considered mentation. Most studies compared programs with “non- in light of these limitations and strengths. The present study active” comparison conditions, for instance where students is among the first umbrella reviews focused on adolescent experience the usual curriculum. This makes it difficult to research, summarizing the state of knowledge in the area of determine in this review whether positive effects relate to secondary school-based mental health interventions (Lev- the components of programs thought to be beneficial for esque, 2016). mental health, such as cognitive restructuring or mindful- ness, or whether positive effect relate to indirect effects like Additional Limitations and Strengths additional one-on-one attention for students from facilita- tors. These sorts of issues related to control conditions are A limitation of the research included in this overview relates already a criticism of depression prevention trials in wider to study settings and the generalizability of findings. Most youth settings. In addition, limited studies included meas- studies in included reviews were conducted in high income ures of positive mental health or wellbeing. The dual con- countries such as Australia, Canada, and the USA, limit- tinua model suggests that positive mental health and mental ing generalizability to other socio-economic and cultural ill-health operate on two spectrums rather than a single con- contexts. Understanding the impact of these differences tinuum (Iasiello et al., 2020). When studies only measure 1 3 Adolescent Research Review one of these spectrums they may fail to provide a complete programs were heterogenous across short and long-term picture of the impact of the intervention. Future studies outcomes. Importantly, the majority of reviews identified could benefit from adding measures of physical, social or for this overview combined findings from primary and sec- educational functioning. It would also be beneficial to draw ondary school settings and age groups, revealing a need upon multiple perspectives by including teachers or parents’ for well-designed systematic reviews that focus on these perspectives on the interventions, as the existing evidence groups separately. The identified gaps in the present review is based primarily on student self-report. Future research can inform future research including systematic reviews, as should aim to tease apart the effectiveness of active program well as recommendations and guidance that can be the basis components to better current understandings around which for improving youth mental health through school-based elements of programs are most effective. interventions. The current overview is also limited by the fact that it Supplementary Information The online version contains supplemen- did not include and compare findings from research in pri- tary material available at https://doi. or g/10. 1007/ s40894- 023- 00211-1 . mary school settings to those from secondary school set- tings. Results presented here provide an umbrella view of Acknowledgements The authors would like to acknowledge all con- tributors to the “Evidence Summary” that inspired this article (see research in secondary school settings, which can be help- reference: Zbukvic et al., 2020). ful for guiding future research and for informing guidance for decision-makers about different program and facilitator Authors’ Contributions IZ conceived of the study, participated in its design and coordination, interpretation of the data and drafted the types. It would be valuable for future reviews to critically manuscript; SM participated in the design and interpretation of the compare and contrast research from across primary, second- data and helped to draft the manuscript; SC participated in the interpre- ary, and indeed tertiary education settings, to fully address tation of the data and helped to draft the manuscript; RA participated developmental and contextual considerations that may influ- in the interpretation of the data and helped to draft the manuscript; VP participated in the interpretation of the data and helped to draft ence program effectiveness. the manuscript; LM participated in the interpretation of the data and A key strength of the present article was the attempt to helped to draft the manuscript; AB participated in the collection and answer a question that had not yet been addressed through interpretation of the data and helped to draft the manuscript; RP par- an overview regarding the effectiveness of school-based pro- ticipated in the interpretation of the data and helped to draft the manu- script; MT participated in the interpretation of the data and coordina- grams targeting anxiety and depression in adolescents. Find- tion helped to draft the manuscript. All authors read and approved the ings are highly relevant to the development of recommen- final manuscript. dations for decision-makers about secondary school-based mental health programs, where guidance is currently limited Funding Open Access funding enabled and organized by CAUL and its Member Institutions. and not always evidence-based. The wide reach of secondary schools globally represents an opportunity for international Declarations collaboration in research and guideline development to sup- port evidence-based decision making in this area. Conflict of interest The authors report no conflict of interest. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- Conclusion tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, Understanding the evidence for secondary school-based pro- provide a link to the Creative Commons licence, and indicate if changes grams targeting anxiety and depression—two of the most were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated common mental health challenges faced by young people— otherwise in a credit line to the material. If material is not included in represents an important area for early intervention. To date, the article's Creative Commons licence and your intended use is not systematic reviews have tended to combine findings from permitted by statutory regulation or exceeds the permitted use, you will research in primary and secondary school settings, produc- need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . ing unclear conclusions about program effectiveness for dif- ferent developmental periods. This overview presents the results of systematic reviews of research focused on second- ary school mental health programs. Overall, the findings are References encouraging and indicate that secondary school settings may be an appropriate context for effective mental health inter - Ahlen, J., Lenhard, F., & Ghaderi, A. (2015). Universal prevention for anxiety and depressive symptoms in children: A meta-anal- ventions targeting depression and anxiety outcomes. Results ysis of randomized and cluster-randomized trials. Journal of showed that targeted interventions for depression and anxi- Primary Prevention, 36(6), 387–403. https:// doi. org/ 10. 1 007/ ety are generally effective in the short term, though evidence s10935- 015- 0405-4 was less clear for long term outcomes; results for universal 1 3 Adolescent Research Review Bajgain, K. T., Amarbayan, M., Wittevrongel, K., McCabe, E., Naqvi, of Child Psychology & Psychiatry & Allied Disciplines, 56(7), S. F., Tang, K., Aghajafari, F., Zwicker, J. D., & Santana, M. 723–734. https:// doi. org/ 10. 1111/ jcpp. 12383 (2023). Patient-reported outcome measures used to improve Das, J. K., Salam, R. A., Lassi, Z. S., Khan, M. N., Mahmood, W., youth mental health services: a systematic review. Journal of Patel, V., & Bhutta, Z. A. (2016). Interventions for adolescent Patient-Reported Outcomes, 7(1), 14. https:// doi. org/ 10. 1186/ mental health: An overview of systematic reviews. Journal of s41687- 023- 00556-0 Adolescent Health, 59(4suppl), S49–S60. h t t p s : / / d o i . o r g / 1 0 . Ballard, M., & Montgomery, P. (2017). Risk of bias in overviews of 1016/j. jadoh ealth. 2016. 06. 020 reviews: A scoping review of methodological guidance and four- De Silva, S., Bailey, A. P., Parker, A. G., Montague, A. E., & Hetrick, item checklist. Research Synthesis Methods, 8(1), 92–108. https:// S. E. (2018). Open-access evidence database of controlled trials doi. org/ 10. 1002/ jrsm. 1229 and systematic reviews in youth mental health. Early Intervention Bastounis, A., Callaghan, P., Banerjee, A., & Michail, M. (2016). The in Psychiatry, 12(3), 474–477. https://doi. or g/10. 1111/ eip. 12423 effectiveness of the Penn Resiliency Programme (PRP) and its Dray, J., Bowman, J., Campbell, E., Freund, M., Wolfenden, L., Hod- adapted versions in reducing depression and anxiety and improv- der, R. K., McElwaine, K., Tremain, D., Bartlem, K., Bailey, J., ing explanatory style: A systematic review and meta-analysis. Small, T., Palazzi, K., Oldmeadow, C., & Wiggers, J. (2017). Journal of Adolescence, 52, 37–48. https://doi. or g/10. 1016/j. adole Systematic review of universal resilience-focused interventions scence. 2016. 07. 004 targeting child and adolescent mental health in the school set- Bell, I., Pot-Kolder, R. M., Wood, S. J., Nelson, B., Acevedo, N., Stain- ting. Journal of the American Academy of Child & Adolescent ton, A., & Allott, K. (2022). Digital technology for addressing Psychiatry, 56(10), 813–824. cognitive impairment in recent-onset psychosis: A perspective. Erford, B., Kress, V., Giguere, M., Cieri, D., & Erford, B. (2015). Schizophrenia Research: Cognition, 28, 100247. Meta-analysis: Counseling outcomes for youth with anxiety dis- Bennett, K., Rhodes, A. E., Duda, S., Cheung, A. H., Manassis, K., orders. Journal of Mental Health Counseling, 37(1), 63–94. Links, P., Mushquash, C., Braunberger, P., Newton, A. S., Kutcher, Feiss, R., Dolinger, S. B., Merritt, M., Reiche, E., Martin, K., Yanes, J. S., Bridge, J. A., Santos, R. G., Manion, I. G., McLennan, J. D., A., Thomas, C. M., & Pangelinan, M. (2019). A systematic review Bagnell, A., Lipman, E., Rice, M., & Szatmari, P. (2015). A and meta-analysis of school-based stress, anxiety, and depression youth suicide prevention plan for Canada: A systematic review of prevention programs for adolescents. Journal of Youth and Ado- reviews. Canadian Journal of Psychiatry, 60(6), 245–257. lescence, 48(9), 1668–1685. Bernecker, S. L., Coyne, A. E., Constantino, M. J., & Ravitz, P. (2017). Ferreira-Vorkapic, C., Feitoza, J. M., Marchioro, M., Simões, J., For whom does interpersonal psychotherapy work? A systematic Kozasa, E., & Telles, S. (2015). Are there benefits from teaching review. Clinical Psychology Review, 56, 82–93. yoga at schools? A systematic review of randomized control trials Bitsko, R. H., Holbrook, J. R., Ghandour, R. M., Blumberg, S. J., Vis- of yoga-based interventions. Evidence-Based Complementary & ser, S. N., Perou, R., & Walkup, J. T. (2018). Epidemiology and Alternative Medicine, 2015, 1. impact of health care provider-diagnosed anxiety and depression Fulambarkar, N., Seo, B., Testerman, A., Rees, M., Bausback, K., & among US children. Journal of Developmental and Behavioral Bunge, E. (2022). Review: Meta-analysis on mindfulness-based Pediatrics, 39(5), 395–403. https:// doi. org/ 10. 1097/ dbp. 00000 interventions for adolescents’ stress, depression, and anxiety in 00000 000571 school settings: A cautionary tale. Child and Adolescent Mental Bradshaw, M., Gericke, H., Coetzee, B. J., Stallard, P., Human, S., & Health. https:// doi. org/ 10. 1111/ camh. 12572 Loades, M. (2021). Universal school-based mental health pro- Galling, B., & Correll, C. (2018). Effectiveness of coordinated spe- grammes in low- and middle-income countries: A systematic cialty care for early psychosis [conference abstract]. Schizophrenia review and narrative synthesis. Preventive Medicine, 143, 106317. Bulletin, 44(suppl1), S108. https://doi. or g/10. 1093/ sc hbul/sb y015. https:// doi. org/ 10. 1016/j. ypmed. 2020. 106317 265 Brown, A., Rice, S. M., Rickwood, D. J., & Parker, A. G. (2016). Gates, M., Gates, A., Pieper, D., Fernandes, R. M., Tricco, A. C., Systematic review of barriers and facilitators to accessing and Moher, D., & Hartling, L. (2022). Reporting guideline for over- engaging with mental health care among at-risk young people. views of reviews of healthcare interventions: Development of the Asia-Pacific Psychiatry, 8(1), 3–22. https://doi. or g/10. 1111/ app y. PRIOR statement. BMJ, 378, 070849. https:// doi. org/ 10. 1136/ 12199bmj- 2022- 070849 Caldwell, D. M., Davies, S. R., Hetrick, S. E., Palmer, J. C., Caro, Gee, B., Reynolds, S., Carroll, B., Orchard, F., Clarke, T., Martin, P., López-López, J. A., & Welton, N. J. (2019). School-based D., & Pass, L. (2020). Practitioner Review: Effectiveness of indi- interventions to prevent anxiety and depression in children and cated school-based interventions for adolescent depression and young people: a systematic review and network meta-analysis. anxiety–a meta-analytic review. Journal of Child Psychology and The Lancet Psychiatry, 6(12), 1011–1020. https://doi. or g/10. 1016/ Psychiatry, 61(7), 739–756. s2215- 0366(19) 30403-1 Hetrick, S., Cox, G., & Merry, S. (2015). Where to go from here? Chahar, A., Gulati, R., Sharma, J., Likhar, N., & Dang, A. (2015). An exploratory meta-analysis of the most promising approaches Impact of community based therapy on children with men- to depression prevention programs for children and adolescents. tal health disorders: A health technology assessment. Value in International Journal of Environmental Research & Public Health, 18(3), 275. Health, 12(5), 4758–4795. https:// doi. org/ 10. 3390/ i jer p h1205 Cordier, R., Speyer, R., Mahoney, N., Arnesen, A., Mjelve, L. H., & 04758 Nyborg, G. (2021). Effects of interventions for social anxiety and Hetrick, S. E., Cox, G. R., Witt, K. G., Bir, J. J., & Merry, S. N. (2016). shyness in school-aged children: A systematic review and meta- Cognitive behavioural therapy (CBT), third-wave CBT and inter- analysis. PLoS ONE, 16(7), e0254117. https:// doi. org/ 10. 1371/ personal therapy (IPT) based interventions for preventing depres- journ al. pone. 02541 17 sion in children and adolescents. Cochrane Database of System- Cowen, S. L. (2014). A meta-analysis of school-based depression pre- atic Reviews. https://doi. or g/10. 1002/ 14651 858. CD003 380. pub4 vention programs for children and adolescents. Brigham Young Hetrick, S. E., O’Connor, D. A., Stavely, H., Hughes, F., Pennell, University. K., Killackey, E., & McGorry, P. D. (2018). Development of an Cristea, I. A., Mogoașe, C., David, D., & Cuijpers, P. (2015). Prac- implementation guide to facilitate the roll-out of early intervention services for psychosis. Early Intervention in Psychiatry, 12(6), titioner Review: Cognitive bias modification for mental health 1100–1111. https:// doi. org/ 10. 1111/ eip. 12420 PMID- 28177 191 problems in children and adolescents: a meta-analysis. Journal 1 3 Adolescent Research Review Hugh-Jones, S., Beckett, S., Tumelty, E., & Mallikarjun, P. (2021). Opinion in Psychiatry, 28(4), 336–342. https:// doi. org/ 10. 1097/ Indicated prevention interventions for anxiety in children and ado-YCO. 00000 00000 000170 lescents: A review and meta-analysis of school-based programs. Oldham-Cooper, R., & Loades, M. (2017). Disorder-specific versus European Child & Adolescent Psychiatry, 30(6), 849–860. generic cognitive-behavioral treatment of anxiety disorders in Hunt, H., Pollock, A., Campbell, P., Estcourt, L., & Brunton, G. children and young people: A systematic narrative review of evi- (2018). An introduction to overviews of reviews: Planning a rel- dence for the effectiveness of disorder-specific CBT compared evant research question and objective for an overview. Systematic with the disorder-generic treatment, Coping Cat. Journal of Child Reviews, 7(39), 1–9. and Adolescent Psychiatric Nursing, 30(1), 6–17. Iasiello, M., van Agteren, J., & Cochrane, E. M. (2020). Mental health Pascoe, M. C., Bailey, A. P., Craike, M., Carter, T., Patten, R. K., and/or mental illness: A scoping review of the evidence and impli- Stepto, N. K., & Parker, A. G. (2021). Single session and short- cations of the dual-continua model of mental health. Evidence term exercise for mental health promotion in tertiary students: A Base, 202(1), 1–45. https:// doi. org/ 10. 21307/ eb- 2020- 001 scoping review. Sports Medicine - Open, 7(1), 72. https://doi. or g/ James, A. C., James, G., Cowdrey, F. A., Soler, A., & Choke, A. (2015). 10. 1186/ s40798- 021- 00358-y Cognitive behavioural therapy for anxiety disorders in children Pearson, M., Chilton, R., Wyatt, K., Abraham, C., Ford, T., Woods, and adolescents. Cochrane Database of Systematic Reviews. H. B., & Anderson, R. (2015). Implementing health promotion https:// doi. org/ 10. 1002/ 14651 858. CD004 690. pub4 programmes in schools: A realist systematic review of research Kambara, K., & Kira, Y. (2021). School-based intervention for depres- and experience in the United Kingdom. Implementation Science, sive symptoms in upper secondary education: A focused review 10, 149. https:// doi. org/ 10. 1186/ s13012- 015- 0338-6 and meta-analysis of the effect of CBT-based prevention. Mental Pfeifer, J. H., & Allen, N. B. (2021). Puberty initiates cascading rela- Health & Prevention. https://doi. or g/10. 1016/j. mhp. 2021. 200206 tionships between neurodevelopmental, social, and internaliz- Karukivi, J., Herrala, O., Sateri, E., Tornivuori, A., Salantera, S., Aro- ing processes across adolescence. Biological Psychiatry, 89(2), maa, M., & Karukivi, M. (2021). The effectiveness of individual 99–108. https:// doi. org/ 10. 1016/j. biops ych. 2020. 09. 002 mental health interventions for depressive, anxiety and conduct Rocha, T.B.-M., Graeff-Martins, A. S., Kieling, C., & Rohde, L. A. disorder symptoms in school environment for adolescents aged (2015). Provision of mental healthcare for children and adoles- 12–18-A systematic review. Front Psychiatry, 12, 779933. https:// cents: A worldwide view. Current Opinion in Psychiatry, 28(4), doi. org/ 10. 3389/ fpsyt. 2021. 779933 330. Labelle, R., Pouliot, L., & Janelle, A. (2015). A systematic review Sancassiani, F., Pintus, E., Holte, A., Paulus, P., Moro, M. F., Cossu, and meta-analysis of cognitive behavioural treatments for suicidal G., Angermeyer, M. C., Carta, M. G., & Lindert, J. (2015). and self-harm behaviours in adolescents. Canadian Psychology/ Enhancing the emotional and social skills of the youth to promote psychologie Canadienne, 56(4), 368–378. https://d oi.o rg/1 0.1 037/ their wellbeing and positive development: A systematic review a0039 159 of universal school-based randomized controlled trials. Clinical Lee, R. S. C., Hermens, D. F., Scott, J., O’Dea, B., Glozier, N., Scott, E. Practice and Epidemiology in Mental Health, 11, 21–40. M., & Hickie, I. B. (2017). A transdiagnostic study of education, Shea, B. J., Reeves, B. C., Wells, G., Thuku, M., Hamel, C., Moran, employment, and training outcomes in young people with mental J., Moher, D., Tugwell, P., Welch, V., Kristjansson, E., & Henry, illness. Psychological Medicine, 47(12), 2061–2070. https:// doi. D. A. (2017). AMSTAR 2: A critical appraisal tool for systematic org/ 10. 1017/ S0033 29171 70004 84 reviews that include randomised or non-randomised studies of Levesque, R. J. R. (2016). Reviews in research on adolescence: Genres, healthcare interventions, or both. BMJ. https:// doi. org/ 10. 1136/ trends, and challenges. Adolescent Research Review, 1(1), 3–13. bmj. j4008 https:// doi. org/ 10. 1007/ s40894- 015- 0022-8 Solmi, M., Radua, J., Olivola, M., Croce, E., Soardo, L., Salazar, G., de Levinson, J., Kohl, K., Baltag, V., & Ross, D. A. (2019). Investigating Pablo, J., Shin, Il., Kirkbride, J. B., Jones, P., Kim, J. H., Kim, J. the effectiveness of school health services delivered by a health Y., Carvalho, A. F., Seeman, M. V., Correll, C. U., & Fusar-Poli, provider: A systematic review of systematic reviews. PLoS ONE, P. (2022). Age at onset of mental disorders worldwide large-scale 14(6), e0212603. https:// doi. org/ 10. 1371/ journ al. pone. 02126 03 meta-analysis of 192 epidemiological studies. Molecular Psychia- Ma, L., Zhang, Y., Huang, C., & Cui, Z. (2020). Resilience-oriented try, 27, 281–295. https:// doi. org/ 10. 1038/ s41380- 021- 01161-7 cognitive behavioral interventions for depressive symptoms in Stockings, E., Hall, W. D., Lynskey, M., Morley, K. I., Reavley, N., children and adolescents: A meta-analytic review. Journal of Strang, J., Patton, G., & Degenhardt, L. (2016). Substance use in Affective Disorders, 270, 150–164. young people 3: Prevention, early intervention, harm reduction, McGorry, P. D., & Mei, C. (2018). Early intervention in youth and treatment of substance use in young people. Lancet Psychia- mental health: Progress and future directions. Evidence Based try, 3(3), 280–296. Mental Health, 21(4), 182. https:// doi. or g/ 10. 1136/ ebmen Townshend, K., Jordan, Z., Stephenson, M., & Tsey, K. (2016). The tal- 2018- 300060 effectiveness of mindful parenting programs in promoting parents’ Meekums, B., Vicky Karkou, E., & Nelson, A. (2015). Dance move- and children’s wellbeing: A systematic review. JBI Database of ment therapy for depression. Cochrane Database of Systematic Systematic Reviews and Implementation Reports, 14(3), 139–178. Reviews. https:// doi. org/ 10. 1002/ 14651 858. CD009 895. pub2https:// doi. org/ 10. 11124/ JBISR IR- 2016- 2314 Moir, T. (2018). Why is implementation science important for interven- UNICEF. (2022). The Global Coalition for Youth Mental Well-being. tion design and evaluation within educational settings? Frontiers UNICEF. Retrieved March 2023 from https:// www. unicef. org/ in Education. https:// doi. org/ 10. 3389/ feduc. 2018. 00061partn ershi ps/ coali tion- youth- mental- wellb eing Nanthakumar, C. (2018). The benefits of yoga in children. Journal Weaver, L. L., & Darragh, A. R. (2015). Systematic review of yoga of Integrative Medicine, 16(1), 14–19. https:// doi. org/ 10. 1016/j. interventions for anxiety reduction among children and adoles- joim. 2017. 12. 008 cents. The American Journal of Occupational Therapy: Official National Academies of Sciences, E., and Medicine. (2019). The prom- Publication of the American Occupational Therapy Association. ise of adolescence realizing opportunity for all youth. National https:// doi. org/ 10. 5014/ ajot. 2015. 020115 Academies Press. https:// doi. org/ 10. 17226/ 25388 Werner-Seidler, A., Huckvale, K., Larsen, M. E., Calear, A. L., Maston, O’Dea, B., Calear, A. L., & Perry, Y. (2015). Is e-health the answer K., Johnston, L., Torok, M., O’Dea, B., Batterham, P. J., Susanne to gaps in adolescent mental health service provision? Current Schweizer, S., Skinner, R., Steinbeck, K., Ratcliffe, J., Oei, J. L., Patton, G., Wong, I., Beames, J., Wong, Q. J. J., Lingam, R., … 1 3 Adolescent Research Review Christensen, H. (2020). A trial protocol for the effectiveness of Younger, D. S. (2016). Epidemiology of childhood and adult mental digital interventions for preventing depression in adolescents: The illness. Neurologic Clinics, 34(4), 1023–1033. https://d oi.o rg/1 0. future proofing study. Trials, 21(1), 2. https:// doi. org/ 10. 1186/ 1016/j. ncl. 2016. 06. 010 s13063- 019- 3901-7 Zalsman, G., Hawton, K., Wasserman, D., van Heeringen, K., Arens- Werner-Seidler, A., Perry, Y., Calear, A. L., Newby, J. M., & Chris- man, E., Sarchiapone, M., Carli, V., Höschl, C., Barzilay, R., Bal- tensen, H. (2017). School-based depression and anxiety preven- azs, J., Purebl, G., Kahn, J. P., Sáiz, P. A., Lipsicas, C. B., Bobes, tion programs for young people: A systematic review and meta- J., Cozman, D., Hegerl, U., & Zohar, J. (2016). Suicide prevention analysis. Clinical Psychology Review, 51, 30–47. https:// doi. org/ strategies revisited: 10-year systematic review. The Lancet. Psy- 10. 1016/j. cpr. 2016. 10. 005 chiatry, 3(7), 646–659. https:// doi. org/ 10. 1016/ S2215- 0366(16) Werner-Seidler, A., Spanos, S., Calear, A. L., Perry, Y., Torok, M., 30030-X O’Dea, B., & Newby, J. M. (2021). School-based depression and Zbukvic, I., Cooke, S., & Anderson, R. (2020). Evidence summary: A anxiety prevention programs: An updated systematic review and review of school-based mental health programs. Orygen. https:// meta-analysis. Clinical Psychology Review, 89, 102079. https:// www. orygen. org. au/ Train ing/ Resou rces/ Depre ssion/ Evide nce- doi. org/ 10. 1016/j. cpr. 2021. 102079summar y/A-r eview-of- secon dar y-sc hool-based- ment al-healt h-p/ WHO. (2012, 2012). Adolescent mental health: mapping actions of orygen- schoo ls- evide nce- summa ry- pdf. aspx? ext= nongovernmental organizations and other international develop- ment organizations. World Health Organization. https://apps. who. Publisher's Note Springer Nature remains neutral with regard to int/ iris/ handle/ 10665/ 44875 jurisdictional claims in published maps and institutional affiliations. Yohannan, J., & Carlson, J. S. (2019). A systematic review of school- based interventions and their outcomes for youth exposed to trau- matic events. Psychology in the Schools, 56(3), 447–464. https:// doi. org/ 10. 1002/ pits. 22202 1 3

Journal

Adolescent Research Review – Springer Journals

Published: Mar 1, 2024

Keywords: Schools; Students; Adolescent; Humans; Mental Health

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Evidence for Targeted and Universal Secondary School-Based Programs for Anxiety and Depression: An Overview of Systematic Reviews

Evidence for Targeted and Universal Secondary School-Based Programs for Anxiety and Depression: An Overview of Systematic Reviews

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Evidence for Targeted and Universal Secondary School-Based Programs for Anxiety and Depression: An Overview of Systematic Reviews

Evidence for Targeted and Universal Secondary School-Based Programs for Anxiety and Depression: An Overview of Systematic Reviews

References (61)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies