School-based versus community-based interventions for promoting mental health among school-age children in marginalised or underserved populations: a systematic review protocol Authors: Nelson Ositadimma Oranye, Asseged Dibaba, Ignatius Uche Nwankwo, Lloyd Webb, Torhonda Lee, Margaret Alexander, Johnpaul Kagulire, Victoria Landu-Adams, Obiajulu Anthony Nnedum, Ehsan Abdalla, Stephen Sodeke, Jonnathan Underwood Published: 2026 ABSTRACT Abstract AbstractIntroductionSchool-aged children frequently experience psychological distress due to academic pressures, a challenge that is often more severe for those from underserved and minority communities. This study aims to evaluate the effectiveness of mental health interventions implemented in school and community settings for children aged 5 to 19. It also seeks to compare the outcomes between children from minority and underserved populations and their peers.Methods and analysisThis systematic review will follow Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to identify relevant studies. Major databases will be searched using a structured search strategy developed by the research team. The review will include randomised controlled trials (RCTs) that assess the impact of interventions conducted in school or community settings to prevent psychological distress—specifically depression, anxiety and stress. To minimise bias, two reviewers will independently select studies and extract data at various stages. The quality of included studies will be assessed. A meta-analysis will be conducted to compare intervention outcomes between children from underserved/minority communities and other children. Pooled prevalence rates and subgroup analyses will be used to explore differences in effectiveness. Heterogeneity among studies and publication bias will also be assessed. Meta-analyses of proportions, ORs and relative risks will be conducted using a random-effects model to estimate effect sizes from multivariate analyses.Ethics and disseminationEthical approval was not required, as this study involved secondary analysis of published literature and did not involve human participants. To date, no systematic review has comprehensively compared school-based and community-based interventions in terms of their effectiveness in addressing anxiety, depression and stress among school-aged children. This review aims to fill that gap by providing clinical insights into the comparative effectiveness of various intervention types and settings.PROSPERO review registrationCRD42023479389.Keywords: Systematic Review, Depression & mood disorders, Anxiety disorders, Physiological Stress, Adolescents STRENGTHS AND LIMITATIONS OF THIS STUDY Few systematic reviews of RCT studies have directly compared the effectiveness of school-based vs community-based mental health interventions for school-aged children. Few systematic reviews of RCT studies have directly compared the effectiveness of school-based vs community-based mental health interventions for school-aged children. The use of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and limitation to RCTs will enhance methodological rigour and internal validity. The use of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and limitation to RCTs will enhance methodological rigour and internal validity. By examining intervention outcomes among children from underserved and minority communities, the study addresses health equity. By examining intervention outcomes among children from underserved and minority communities, the study addresses health equity. However, limiting inclusion to RCTs may have excluded valuable evidence from high-quality quasi-experimental or real-world implementation studies. But this was intended to focus the study, considering the extensive studies in the field. However, limiting inclusion to RCTs may have excluded valuable evidence from high-quality quasi-experimental or real-world implementation studies. But this was intended to focus the study, considering the extensive studies in the field. Potential publication bias exists because unpublished studies or studies with null findings may be under-represented. Potential publication bias exists because unpublished studies or studies with null findings may be under-represented. Introduction Anxiety, stress and depression are serious mental health conditions that frequently begin in early childhood and can have significant and lasting effects on the well-being of school-aged children and adolescents.1 2 Compared with adults, young people are at a higher risk of developing mental health disorders,3,6 with approximately 50% of such conditions beginning before the age of 14.7 8 This elevated risk is believed to stem from the challenges associated with growing up, underscoring the importance of early screening and intervention to reduce long-term effects and disabilities associated with mental health disorders.8 The COVID-19 outbreak in 2019/2020 has been linked to a surge in mental health issues across the general population, with particularly high prevalence rates observed among school-aged children.9 10 The COVID-19 pandemic is involved in enhancing anxiety, stress and depression.11,13 Importantly, the COVID-19 pandemic increased the burden and disability related to anxiety, stress and depression in terms of enhanced vulnerability of specific subgroups of subjects.14 The individual well-being and quality of life have been significantly related to increased psychosocial issues during COVID-19, and this has enhanced the vulnerability of patients, particularly youths who are very vulnerable to stressful situations.15 Accordingly, this review will focus on the COVID-19 and post-COVID-19 periods. Psychological distress in this study is broadly defined as ‘a state of emotional suffering’,16 encompassing conditions such as depression, anxiety and stress. Globally, an estimated 10%–20% of adolescents experience mental health conditions,17 while in the USA, around 10%–12% of children are affected by anxiety and depression.18,20 UNICEF’s 2020 ranking of child wellness,21 which includes mental well-being, placed the USA near the bottom among developed countries, 36th out of 38, highlighting poor outcomes in child well-being. Although numerous studies have reported elevated levels of psychological distress in the general population,10 22 schoolchildren often experience distress related to educational demands.123,25 Beyond the developmental challenges of adolescence, educational pressure appears to be a significant risk factor contributing to increased psychological distress among school-aged children. Rates of anxiety and depression also vary according to social characteristics and backgrounds, including early childhood experiences, gender and life circumstances.1 21 26 Gender, in particular, has emerged as an important variable in mental health disparities. Some studies suggest that male university students reported better well-being than female students during the COVID-19 pandemic, with socioeconomic status playing a role in these differences.27 However, findings on gender differences in mental health outcomes have been inconsistent. For example, while UNICEF reported higher rates of mental health issues among males, other studies have indicated that females may be at greater risk. A systematic review by Metin et al26 found greater anxiety and fear among male university students, whereas broader population studies have yielded mixed or contradictory results.25 28 29 These findings illustrate the complexity of how various social factors, including gender, may influence or modify the mental well-being of school-aged children. Comorbidity of mental illness Adolescents with pre-existing mental health conditions face significantly greater challenges,13 30 31 highlighting the importance of addressing comorbidity in child and adolescent mental well-being. The link between mental illness and comorbid conditions became particularly evident during the COVID-19 pandemic. Following the WHO’s declaration of COVID-19 as a global pandemic in March 2020, widespread lockdowns were implemented, lasting until around August 2020.32 33 These lockdowns, coupled with the extended school closures, were strongly associated with heightened psychological problems, such as anxiety and stress among children and adolescents.6 12 34 35 While the pandemic led to a general rise in anxiety and stress across the population,9 the prevalence was notably higher among school-aged children and adolescents.6 36 37 Psychological distress tends to escalate during periods of social upheaval, as was observed during the pandemic. However, limited attention has been given to the specific psychological impact of the broader social disruption caused by COVID-19 on adolescent mental health.38 Our understanding of how different types of mental health interventions, such as school-based versus community-based, compare in terms of effectiveness, particularly among vulnerable populations, remains limited due to the multifaceted nature of contributing factors. Race The relationships between race, racism and perceived discrimination and their associations with stress, anxiety and depression in youth and adults are well documented in the literature.6 39 While depression, anxiety and stress are recognised as significant mental health concerns among young people in general, racialised minority youth are at greater risk of developing these conditions.14 40 41 Research has consistently shown that children from racialised and marginalised communities bear a disproportionate burden of mental health problems compared with their peers from more affluent groups.39 42 43 Although increased levels of anxiety, stress and depression were reported among children and adolescents during the COVID-19 pandemic,6 37 the pandemic disproportionately affected Black and other marginalised communities. For example, Lundberg et al44 reported significantly higher COVID-19 death tolls among Black Americans and Latinos compared with White and Asian counterparts. These elevated infection and mortality rates likely contributed to widespread psychological trauma among children and adolescents. Chen et al36 found that individuals of African and West Asian descent experienced worse mental health outcomes during the pandemic. These results aligned with arguments by Page et al,45 suggesting that racial and ethnic minority groups, especially those already facing systemic inequities and economic hardship, were more vulnerable to higher infection and mortality rates, with long-lasting psychological consequences. Global research corroborates this, showing that ethnic minority groups in the UK were more severely affected than their White British counterparts.46,48 Despite growing research linking racism and perceived discrimination to anxiety and depression,39 there is still a lack of systematic reviews of RCTs evaluating the effectiveness of mental health interventions for vulnerable children from underserved and racialised communities. There is also a pressing need for reviews assessing mental health interventions in the context of the COVID-19 pandemic and its impact on marginalised populations. Social context Schoolchildren come from diverse social backgrounds, including socioeconomic status, gender and racial or ethnic identity, which influence their psychosocial health.26 42 49 Numerous studies have highlighted the impact of these social contexts on mental well-being25 27 50 and the effectiveness of interventions, particularly among underserved and minority schoolchildren.42 43 49 Children from marginalised communities often face additional layers of psychological stress due to compounding social factors such as poverty, racism and systemic discrimination. These sociogenic stressors exacerbate existing disparities in health outcomes. It is essential to consider the broader social context when understanding how mental health challenges affect different population groups. Evidence from the COVID-19 pandemic underscores the role of social determinants in shaping mental health outcomes. Minority populations were disproportionately impacted due to structural inequalities, social exclusion and resource limitations.46 51 Children and adolescents from racialised communities experienced higher infection rates and greater psychological distress.45 Comorbidities also played a critical role. Adolescents with pre-existing mental health conditions were especially vulnerable during the pandemic.13 30 31 These intersecting factors reinforce the need for targeted and nuanced mental health interventions. A systematic review that accounts for the differential burden of mental health issues among marginalised children is critically important. Such a review can identify best practices in both school and community settings and inform interventions that are more effective for underserved populations. Access to mental health A major contributor to mental health disparities is the lack of access to adequate healthcare services in minority and underserved communities. Studies have shown that mental health challenges are often worsened by systemic disparities in access to care, particularly among low-income, racial and ethnic minority youth.44 52 53 In areas lacking hospitals or major healthcare infrastructure, it is estimated that between 44% and 70% of individuals who need mental health interventions do not receive appropriate care.54 55 Stigma is another significant barrier. As former US Surgeon General David Satcher emphasised, stigma may be the greatest obstacle to mental healthcare access.56 57 School-based mental health services have shown promise in improving psychological outcomes and educational performance.55 However, these interventions often exclude parental involvement due to logistical challenges. This exclusion is problematic, as parental participation can offer valuable insights and support.58 Community-based interventions present a promising alternative, offering a different social environment that may yield unique therapeutic benefits. Non-traditional settings such as schools and community organisations can be more effective in reaching vulnerable populations, often being more accessible and less stigmatised than specialised clinics.55 59 This review will compare the effectiveness of school-based versus community-based interventions by examining subgroup differences across demographic and social factors, including race, ethnicity, geographic location, gender, age and socioeconomic status. The analysis will assess the differential effects on children from Black and minority communities, with data stratified into COVID-19 and post-COVID-19 periods. Theoretical framework This study is grounded in social theories related to relative deprivation, racial inequality and health disparities.43 60 61 Insights gained from the epidemiology of COVID-19 underscore the importance of incorporating social context into the health intervention design. For instance, Lemyre et al27 demonstrated that gender differences in well-being during the pandemic were strongly associated with socioeconomic status. Other studies found that adolescents with pre-existing conditions experienced more severe challenges13 30 31 and that race, ethnicity and minority status significantly shape mental health outcomes.42 43 49 Depression is frequently identified as a comorbidity of anxiety, which is why this study jointly examines anxiety, depression and stress. These interconnected conditions require a holistic approach that considers both clinical and social determinants of health. Contextual gaps in existing systematic reviews Although recent reviews62 63 examined outcomes related to depression and anxiety, their primary emphasis was on broader mental health promotion and prevention efforts delivered exclusively through school-based programmes. Generally, existing systematic review evidence is disproportionately concentrated on school-based mental health interventions,64,67 revealing a substantial gap in the evaluation of community-based approaches and a complete absence of direct comparisons between school and community settings in underserved rural populations. Other reviews68,71 focused almost exclusively on school-delivered interventions, primarily in low- and middle-income countries (LMICs) or socioeconomically disadvantaged settings, thereby reinforcing a predominantly school-centric evidence base. Even where reviews incorporated both school- and community-based interventions,71 72 differences in effectiveness by delivery setting were not examined. Hudson et al73 similarly synthesised evidence across school, community and clinical settings without disaggregating findings by intervention context. Collectively, the predominance of RCTs—sometimes supplemented with quasi-experimental designs—restricts applicability to real-world rural and community settings, underscoring a critical need for comparative and context-sensitive effectiveness research. Despite incorporating evidence from LMICs and socioeconomically disadvantaged populations, existing reviews did not specifically focus on or stratify results for marginalised children in underserved rural communities. Analyses were typically conducted at an aggregate LMIC or general school-population level, with rurality and marginalisation treated as implicit rather than analytically distinct contextual factors.74 As a result, despite robust evidence supporting school-based mental health interventions and increasing inclusion of community-based approaches, no systematic review has directly compared the effectiveness of school- versus community-based interventions for children in underserved rural settings. This omission represents a critical gap in the literature, limiting the evidence available to inform equitable, context-appropriate service delivery in rural contexts. This review targeted adolescents and school-aged children, whereas several existing systematic reviews focus on higher-education or college student populations,24 27 75 76 reducing the relevance of their findings for younger populations. Justification for the review To date, no systematic review has comprehensively examined the comparative effectiveness of school-based versus community-based mental health interventions for children and adolescents. Existing reviews have largely considered these delivery settings in isolation, limiting their utility for informing context-sensitive intervention selection. In addition, this review prioritised clinically relevant primary outcomes—depression, anxiety and stress—that are highly prevalent during childhood and adolescence yet inconsistently synthesised across intervention settings in the current literature. Several systematic reviews have focused more on general mental health or prevalence studies than intervention outcomes,22 28 77 while some focused on specific mental health problems like suicidal ideation.78 There is also a lack of research exploring how outcomes differ between youth from minority and underserved communities and those from more advantaged backgrounds. Furthermore, existing studies on psychological issues in minority groups often yield inconsistent findings, reflecting variation in intervention design and implementation. These differences can significantly influence treatment outcomes. This review will help bridge knowledge gaps by identifying effective, evidence-based practices. Insights will inform the development of clinical tools and guidelines to support mental health promotion in educational and community settings. Ultimately, the review will benefit schools, parents and organisations working to enhance child and adolescent mental health globally. Methods This protocol follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 Guidelines79 and focuses on RCTs evaluating programmes aimed at preventing anxiety, depression or stress in school-aged children. A sample PRISMA 2020 checklist is provided in online supplemental appendix A. The protocol is registered with PROSPERO (CRD42023479389). Search strategy The search strategy was developed with health science librarians using a PICO framework. Databases include PubMed, ScienceDirect, ProQuest PsycINFO, Ovid Global Health and grey literature. The review will include studies published from 2000 onwards. Core search terms include school-based intervention, community-based intervention, depression, anxiety, stress, promoting mental health, mental well-being, school-age children, adolescents, underserved, marginalised, health disparities, minority group and RCT. These terms were tested and refined through multiple iterations. Boolean operators and exclusion terms (eg, adult, college) were used to refine results. An example of a search map used is (“School-Based Intervention” OR school-based OR school intervention OR community-based OR “Community Health Services”) AND (intervention OR programme OR prevention) AND (“Mental Health” OR “Mental Well-Being” OR depression OR anxiety OR stress, psychological) AND (“Child” OR “Adolescent” OR school-age child OR teen) AND (underserved OR marginalised OR “Health Disparities” OR minority group OR low-income) AND (“Randomised Controlled Trial” Publication Type OR randomised) NOT (adult OR college OR university OR higher education) Filters: Publication date from 1/1/2000–21/12/2025. Database filters were applied to restrict results to original research articles and to exclude review publications, including scoping and systematic reviews. The database search strategies are provided in the online supplemental appendix B. Review articles that nonetheless appeared in search results were removed during the Level 1 screening (title and abstract review). See online supplemental appendix C for example of articles excluded. Grey literature will be sourced through Google, Google Scholar and bibliographies. Expert input will be obtained from academic researchers and clinical practitioners.80 Auto-alert in Ovid and PubMed Central will notify the team of new studies. The first 200 Google Scholar results will be reviewed to ensure that potentially eligible studies are not missed. English-language articles will be prioritised, with French-language studies considered if team members are proficient. Reference lists of relevant reviews will also be examined for eligible studies not captured in the initial search. Study participants The target population includes school-aged children aged 5–19, consistent with the US Department of Education’s National Center for Education Statistics definition.81 Inclusion and exclusion criteria To be eligible for inclusion, studies must use an RCT design and measure depression, anxiety or stress as primary outcomes, assessed at baseline and postintervention or across multiple time points. Additionally, eligible studies must involve interventions delivered specifically in school or community/home settings. Studies that do not meet these criteria will be excluded. Furthermore, interventions conducted in clinical or hospital settings as part of routine medical care, as well as non-RCT studies or those that do not report relevant mental health outcomes, will not be included in the review. Table 1 summarises the eligibility criteria for the studies included in this review. Table 1Eligibility criteria for this systematic reviewPICO acronymInclusion criteriaExclusion criteriaP—participant/populationChildren, adolescents 5–19 years of age, in school or out of school (in community setting)>19 years old, clinical populations, non-school or community context (clinical or institutional population, in any form of institutional confinement like penitentiary)I—interventionIntervention or training to prevent or promote mental health or mental well-being (specifically Depression, Anxiety or Stress, DAS)No explicit intervention implemented targeting DASC—control/comparisonControl group that does not complete the intervention being implementedNo control/comparison groupO—outcomesAt least one of the mental well-being-related outcomes of DAS. Related outcomes of mental well-being will be recorded, but only in studies that include at least one of the DAS.Studies that do not report on participant outcomesStudies with no measured mental well-being related to the outcomes of DASAdditional criteriaRandomised controlled trials.(Keep in view=quasi-randomised trials, non-randomised trials and experimental studies).Correlational, cross-sectional and qualitative studies; literature reviews and case reportsNon-English language articles Eligibility criteria for this systematic review To minimise heterogeneity among the included studies, those employing standardised intervention protocols will be considered and quality of RCT design will be evaluated, as recommended in previous research.36 82 Outcomes of interest The primary outcomes of interest in this review are depression, anxiety and stress. These outcomes will be the focus for evaluating the effectiveness of the interventions. In addition to these, secondary outcomes related to psychological distress as defined within the individual studies included in the review will also be considered. While depression, anxiety and stress are the core outcomes, the review will also extract and analyse contextual factors, including intervention setting, school type, population characteristics (eg, underserved or minority groups) and the nature of the intervention (eg, behavioural, physical activity–based or educational). Study selection Study selection and data extraction will be conducted by two independent reviewers. To minimise the risk of bias, reviewers will be paired at different stages of the process, such as the Level 2 full-text review and data extraction based on their expertise and experience. This means the same pair of reviewers will not assess the same studies at every stage. Each reviewer will independently evaluate a set of assigned articles using the previously defined inclusion and exclusion criteria. In cases where the two reviewers disagree on the eligibility of a study, they will first attempt to resolve the conflict through discussion. If a consensus cannot be reached, a third reviewer will be brought in to resolve the disagreement. Review management and data extraction The review process will be managed using DistillerSR software, which supports automation of literature screening, data extraction and reporting through artificial intelligence and intelligent workflows.63 Data extraction will be performed using a standardised template provided within DistillerSR. For researchers who may need to work offline, like international collaborators, an Excel-based backup template will be used. Only studies employing RCT designs to evaluate the effectiveness of interventions on mental health outcomes in children aged 5 to 19 years will be included in the extraction process. Two independent researchers will extract data, and all entries will be cross-checked by the principal investigator. In cases of discrepancies, the original reviewers will be asked to confer. If they cannot reconcile their differences, a third reviewer will be appointed to independently assess the case. All decisions, reviewer comments and extracted data will be documented in DistillerSR and can be exported into Excel or other formats as needed. Outcomes will be extracted in both absolute terms (eg, prevalence or rate) and relative terms (eg, OR, relative risk, risk difference or rate ratio). Demographic information, including gender/sex, age, socioeconomic status and minority or underserved status (if reported), will also be collected. In instances where critical outcome data appear to be missing, study investigators will be contacted to clarify whether the data were collected or reported elsewhere. Strategy for data synthesis The data synthesis process will be supported by the Distiller Systematic Review Management software,83 which includes built-in capabilities to compute inter-rater statistics such as percentage agreement and inter-rater reliability. These metrics, along with other relevant psychometric statistics related to reviewer performance, will be documented and reported. The methodological quality of the included studies will be assessed using two established tools: the Cochrane Collaboration’s tool for assessing risk of bias84 and the Jadad scale. To compare outcomes for children from underserved and minority communities with those from more socially advantaged groups, a meta-analysis will be conducted using pooled prevalence estimates and subgroup analyses. Stratified analyses will also be conducted to examine the influence of additional variables, such as gender, academic programme type (eg, Science, Technology, Engineering, and Mathematics (STEM)) and other relevant characteristics when available. Data from comparable studies will be pooled to assess the impact of specific types of interventions and settings (eg, school-based vs community-based), but only when the interventions are deemed sufficiently similar and the studies report or allow for the calculation of effect sizes and CIs. Subgroup analyses will be conducted with a 95% CI to further explore heterogeneity across studies. The degree of heterogeneity will be quantified using the Cochran Q statistic and the I² test.22 61 Publication bias will be assessed using Egger’s and Begg’s tests.85,87 Where sufficient quantitative data from multivariate analyses are available, a meta-analysis will be performed to compute proportions, adjusted ORs and relative risks, using a random effects model to account for variability between studies. In cases where quantitative data are insufficient or too heterogeneous for statistical pooling, a meta-synthesis will be conducted to qualitatively integrate the findings. Status and timeline of the review This review is currently ongoing and has advanced to the Level 3 data extraction by conducting data extraction using templates developed in the DistillerSR platform. The team aims to complete the data extraction and analysis within 2–3 months. The primary hypothesis is that the COVID-19 pandemic disproportionately exacerbated symptoms of depression, anxiety and stress among children living in low-resource communities compared with their peers in high-resource communities. A secondary hypothesis is that community-based interventions involving parents or guardians are more effective in reducing depression, anxiety and stress than school-based interventions alone. Patient and public involvement None. However, we plan to apply for a knowledge translation and exchange grant to actively engage community networks and affected populations in rural and remote communities throughout the translation process. Community action agencies, such as schools, parents’ associations and youth societies, will be involved in disseminating study findings to broader patient and community audiences. These stakeholders will contribute to decisions regarding which information and results are shared, the timing of dissemination and the most appropriate formats for communication. Supplementary material Acknowledgements The research team is very grateful to the Tuskegee University RCMI leadership team, Dr Deepa Bedi, Dr Temesgen Samuel and Dr Timothy Turner, for supporting and encouraging us to complete this project. This project would not have been successful without their inspiration and trust in our team. We also want to thank the Tuskegee University RCMI staff, Chiquita Lee, Lecarde Webb and Kamilah Williams. Footnotes Funding: This research was supported through the Tuskegee University CBR/RCMI NIH grant #U54MD007585 in 2024/2025 pilot grant cycle. The funder had no role in the design of the study; the collection, analysis, or interpretation of data; the writing of the manuscript or the decision to submit the manuscript for publication. The results and outcomes of the study were not influenced by the funder, despite author affiliations. Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-113697). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Patient and public involvement: Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research. References Supplementary Materials