Heterogeneity among modes of interventions

The findings revealed distinct heterogeneity in the modes of intervention delivery, namely physical and digital/technology-based approaches. Digital/ technology-based interventions were noted to be prominent in high-income countries, with 12 interventions reported, compared to five in upper-middle-income countries and two in low- and lower-middle-income countries. This disparity underscored the reliance of such interventions on robust technological infrastructure, such as widespread internet connectivity (WHO, 2022), high smartphone penetration rates (International Telecommunication Union, 2023) and the availability of advanced technological resources (Chakrabarti, Reference Chakrabarti2024), which are more prevalent in high-income settings.

Digital/ technology-based interventions resonate with youth due to their interactive and engaging nature (Ferretti et al., Reference Ferretti, Hubbs and Vayena2023). In contrast, the minimal presence of digital interventions in low- and lower-middle-income countries highlighted significant barriers, including limited access to digital technology, high costs associated with devices and data and infrastructural challenges (WHO, 2020). More physical interventions, by contrast, were used in low- and lower-middle-income countries. These findings emphasised the continued importance of physical approaches in regions where technological barriers restrict the feasibility of digital solutions. As such, physical interventions may offer greater adaptability and cultural appropriateness to local contexts, considering their non-reliance on technological infrastructure.

The disparities in intervention types highlighted the importance to understand that digital health should not be regarded as a panacea, which calls for balanced and context-sensitive approaches to mental health improvement. Each modality holds distinct value and should complement, not replace, the other. While digital interventions can offer scalability, accessibility and cost-effectiveness particularly in reaching remote or underserved populations (Schueller et al., Reference Schueller, Hunter, Figueroa and Aguilera2019), physical interventions offer personal interaction, immediate feedback, and may be more suitable for complex cases or in contexts where digital literacy is low. This diversity also suggested the potential for integrating digital tools with physical approaches to optimise intervention effectiveness, particularly in transitional settings (WHO, 2020). For instance, while physical interventions may serve as the primary mode in low-income countries, the gradual incorporation of digital elements could extend reach and scalability as technological infrastructure evolves. This hybrid approach could leverage the strengths of both physical and digital interventions, potentially overcoming the limitations of each when used in isolation (Nahum-Shani et al., Reference Nahum-Shani, Dziak, Walton and Dempsey2022), accommodating diverse needs and settings, as well as ensuring equitable access to mental health care.

Heterogeneity among intervention modalities and settings

This review also revealed a notable trend in the intervention modalities for youth aged 15 to 24, with CBT, psychoeducation and mindfulness-based interventions emerging as the most prominently used approaches compared to other modalities. This could potentially suggest that CBT’s structured nature and emphasis on skill-building (Nelson & Tusaie, Reference Nelson and Tusaie2011), psychoeducation’s emphasis on provision of both disease-specific information (e.g., recognition and management of disorders) and general information (e.g., promotion of optimal well-being) (Bjørnsen et al., Reference Bjørnsen, Eilertsen, Ringdal, Espnes and Moksnes2017), and mindfulness’ key feature of paying attention to the events occurring within one’s mind and body (Napoli et al., Reference Napoli, Krech and Holley2005) align well with the developmental needs of youths, who are in a critical period of cognitive and emotional growth. These approaches have consistently demonstrated strong empirical effectiveness across a wide range of mental health issues common in this age group, such as anxiety, depression and stress-related disorders, attention deficit and hyperactivity disorder, conduct disorder and behavioural addictions (Semple et al., Reference Semple, Lee and Miller2006; Hofmann et al., Reference Hofmann, Asnaani, Vonk, Sawyer and Fang2012; Banaschewski et al., Reference Banaschewski, Becker, Döpfner, Holtmann, Rösler and Romanos2017; Jones et al., Reference Jones, Thapar, Stone, Thapar, Jones, Smith and Simpson2017; Cardamone-Breen et al., Reference Cardamone-Breen, Jorm, Lawrence, Rapee, Mackinnon and Yap2018; González-Bueso et al., Reference González-Bueso, Santamaría, Fernández, Merino, Montero, Jiménez-Murcia, Del Pino-Gutiérrez and Ribas2018).

While these approaches address the cognitive-behavioural aspects of mental health and the broader understanding of mental well-being, most included studies did not report culturally adapted interventions, raising questions about the need for more diverse interventions tailored to specific cultural contexts, individual preferences and varying mental health needs of this age group. The need for culturally adapted interventions stems from the recognition that many mental health interventions, including CBT, are often developed within Western cultural frameworks (Naeem et al., Reference Naeem, Sajid, Naz and Phiri2023), which may not fully align with the values, beliefs and lived experiences of individuals from non-Western or minority cultural backgrounds. Research has shown that cultural factors significantly influence the acceptability, engagement and effectiveness of interventions. For instance, studies have demonstrated that some principles underlying CBT can conflict with cultural beliefs and practices, resulting in lower recruitment, retention and engagement among ethnic minority participants (Naeem et al., Reference Naeem, Phiri, Rathod and Ayub2019; Rathod et al., Reference Rathod, Phiri and Naeem2019).

Although this review did not explicitly analyse the outcomes of culturally adapted versus non-adapted interventions, existing literature provides compelling evidence for the importance of cultural tailoring. For example, culturally adapted CBT has been shown to improve access, engagement and outcomes for diverse populations (Naeem et al., Reference Naeem, Phiri, Rathod and Ayub2019; Naeem et al., Reference Naeem, Sajid, Naz and Phiri2023). Frameworks for cultural adaptation emphasise the need to address barriers such as language, cultural stigma and differing conceptualisations of mental health. These findings highlight the importance of tailoring interventions to ensure their relevance and effectiveness across diverse settings.

This underscores the importance of flexibility and individualisation in mental health care. Rather than adhering to a one-size-fits-all approach, interventions should be adapted to meet the unique needs, preferences and circumstances of each individual (Stanhope et al., Reference Stanhope, Ingoglia, Schmelter and Marcus2013; Miller et al., Reference Miller, Stanhope, Restrepo-Toro and Tondora2017; Moreno & Chhatwal, Reference Moreno and Chhatwal2020). This recognition of diversity within the treatment approaches signifies a shift towards holistic and inclusive models of care, prioritising symptom reduction alongside overall well-being and resilience (Cloninger et al., Reference Cloninger2010). This would potentially expand the width of exploration and innovation in mental health research and practice, ultimately enhancing outcomes for youths facing mental health challenges.

In terms of delivery platforms, educational settings emerged as the most common, offering accessibility, built-in populations and opportunities for stigma reduction (Fazel et al., Reference Fazel, Hoagwood, Stephan and Ford2014; O’Reilly et al., Reference O’Reilly, Dogra, Whiteman, Hughes, Eruyar and Reilly2018). Schools facilitate consistent engagement and leverage existing infrastructure (Weist et al., Reference Weist, Mellin, Chambers, Lever, Haber and Blaber2012), making them ideal venues for universal and preventive efforts. However, over-reliance on school-based settings risks excluding out-of-school youth such as migrants, homeless individuals or young parents, who often face heightened vulnerability (UNICEF & UNESCO Institute for Statistics, 2016). Community and healthcare settings, though less frequently used, played important roles in delivering targeted or clinical interventions, often functioning as referral points from schools. Meanwhile, general population platforms particularly digital and social media remain underused despite growing evidence that youth value the anonymity, flexibility and immediacy offered by such channels (Rickwood et al., Reference Rickwood, Deane and Wilson2007; Pretorius et al., Reference Pretorius, Chambers and Coyle2019). Their limited uptake may be attributed to their relative novelty, as well as the ongoing need to validate their safety, efficacy and ethical deployment.

Collectively, these findings underscore the importance of diversifying both intervention modalities and delivery settings. Expanding beyond traditional school-based platforms to include community, clinical and digital approaches particularly those grounded in youth-aligned and culturally sensitive frameworks can help bridge current implementation gaps. A more balanced and inclusive implementation strategy is needed to reach underserved and high-risk youth, particularly those outside formal education systems, thereby promoting equity and effectiveness in youth mental health support (Naslund et al., Reference Naslund, Aschbrenner, Araya, Marsch, Unützer, Patel and Bartels2017).

Heterogeneity of mental health concepts

The systematic review had also revealed that interest in youth mental health interventions has grown steadily over the past decade, with a particularly notable increase in 2019. This signified the global awareness and urgency to explore both mental health symptoms and well-being outcomes among youth aged 15 to 24, which also aligned with the emergence of key statistics and research emphasising its significance (e.g., WHO, 2019; UNICEF, 2021). More importantly, this review uncovered significant diversity within the mental health concept, notably distinguishing between mental health symptoms viewed through the lens of disorder and mental well-being from a transdiagnostic perspective. Specifically, 38 studies focused on specific mental health symptoms such as anxiety, depression and suicidal tendencies, while 25 studies targeted mental well-being, encompassing aspects like emotional regulation, self-efficacy, self-esteem and daily functioning. Aligning with the current categorisation of youth into General Youth, Vulnerable Youth, Youth at Risk and Youth with Diagnoses, this suggests a growing urgency of mental health continuum from normative stressors to clinically significant conditions (Kessler et al., Reference Kessler, Chiu, Demler and Walters2005; Patel et al., Reference Patel, Flisher, Hetrick and McGorry2007), as well as the understanding of the multifaceted nature of mental health and the necessity for comprehensive approaches to support individuals’ overall mental status.

Throughout history, distress and suffering have been recognised as inherent aspects of the human condition (Kendler, Reference Kendler2009). The interpretation and characterisation of mental distress have evolved significantly, particularly in Western contexts, where mental health struggles have traditionally been categorised within formal taxonomic systems like the DSM and the International Classification of Diseases (Kendler, Reference Kendler2009). These systems serve as foundational frameworks in psychiatry, clinical psychology and abnormal psychology, guiding training across various helping professions and shaping global approaches to assessing, managing and treating mental health issues (Cosgrove et al., Reference Cosgrove, Krimsky, Vijayaraghavan and Schneider2006; Marecek and Hare-Mustin, Reference Marecek, Hare-Mustin, Fox, Prilleltensky and Austin2009). However, there is a growing recognition of the limitations of these taxonomies, including issues of heterogeneity within diagnoses and incomplete symptom capture (Dalgleish et al., Reference Dalgleish, Black, Johnston and Bevan2020), leading to the emergence of transdiagnostic approach.

Transdiagnostic approach diversifies the perspectives on understanding mental health challenges and fosters new conceptualisations of the underlying theories and processes of mental distress. At its core is the recognition that risk, protective and maintenance factors and processes implicated in mental health problems, whether biological, socio-environmental or psychological, do not exhibit specificity for particular diagnostic disorders but instead operate across traditional diagnostic boundaries (Buckholtz & Meyer-Lindenberg, Reference Buckholtz and Meyer-Lindenberg2012). This approach has been shown to have an impact on assessment (Mansell, Reference Mansell2019), treatment and recovery (Murray et al., Reference Murray, Metz and Callaway2019). This notion is also supported by a body of research (e.g., Haslam et al., Reference Haslam, Holland and Kuppens2011; Waszczuk et al., Reference Waszczuk, Zimmerman, Ruggero, Li, MacNamara, Weinberg, Hajcak, Watson and Kotov2017) where mental health problems are most effectively understood along continua rather than as discrete entities. However, like the disorder perspective, the transdiagnostic approach also has limitations, where one significant drawback is its lack of a clear phenomenological framework, which can be ambiguous and lacking specificity (Fusar-Poli et al., Reference Fusar-Poli, Solmi, Brondino, Davies, Chae, Politi, Borgwardt, Lawrie, Parnas and McGuire2019). This ambiguity could complicate the measurement of psychometric items in non-clinical samples that may exhibit clinical symptoms or established mental disorders.

Considering the advantages and limitations of both approaches, it is vital to consider the proposition that mental health is not simply a uniform state of being, such as depression, anxiety or delusion (Parnas, Reference Parnas2015), nor is it solely the result of an array of subjective appraisals of problems (Ghaemi, Reference Ghaemi2006; Parnas, Reference Parnas2014). Instead, it is the integration of both diagnostic and transdiagnostic concepts where mental health is to be perceived as a multifaceted construct encompassing diverse dimensions. It is consistent with the concept where humans navigate a continuum, encountering a spectrum of symptoms ranging from mild distress to severe psychiatric disorders, each characterised by distinct origins, presentations and impacts on daily functioning (Keyes, Reference Keyes2002). As such, there is value in recognising both an individual’s position on this continuum as posited in the transdiagnostic approach and the point at which certain difficulties persist beyond a potentially individualistic threshold, thereby constituting disorders based on the diagnostic approach. This notion is consistent with the hierarchical framework as exemplified by the Hierarchical Taxonomy of Psychopathology (HiToP; Kotov et al., Reference Kotov, Krueger, Watson, Achenbach, Althoff, Bagby, Brown, Carpenter, Caspi, Clark, Eaton, Forbes, Forbush, Goldberg, Hasin, Hyman, Ivanova, Lynam, Markon and Miller2017), where it organises mental health problems into a hierarchy, with a general factor of psychopathology at the apex, broad spectra like internalising and disinhibited/externalising problems beneath, and progressively lower levels detailing co-occurring disorders, symptom components and individual signs. This perspective holds weight in assessment, diagnosis, treatment and management of mental health issues, providing a more comprehensive framework for addressing the complexities of mental well-being (Dalgleish et al., Reference Dalgleish, Black, Johnston and Bevan2020).

Heterogeneity among measurement tools

The diverse conceptualisations of mental health across studies naturally led to a variety of measurement tools, each designed to capture specific facets of its complex and multifaceted nature. For instance, depressive symptoms were assessed using standardised tools such as the 9-item Patient Health Questionnaire (PHQ-9), Center for Epidemiological Studies-Depression Scale, Mood and Feelings Questionnaire and Beck Depression Inventory-II (BDI-II). These instruments, including the BDI (Beck et al., Reference Beck, Steer and Brown1996) and the PHQ (Kroenke et al., Reference Kroenke, Spitzer and Williams1999), provided standardised scoring systems and cut-offs, facilitating comparisons across studies and populations. Similarly, the Generalized Anxiety Disorder Scale (Spitzer et al., Reference Spitzer, Kroenke, Williams and Löwe2006) and the State–Trait Anxiety Inventory offered standardised assessments of anxiety symptoms, enabling researchers to quantify and track changes over time. The current difference in measurement outcome, prevalence rate and sensitivity of scores could give rise to further heterogeneity in interpretation, which highlights the importance of standardised instruments.

Standardised instruments play a crucial role in mental health research by providing a common language for assessment and enabling objective and quantitative evaluation of psychopathological phenomena (Möller, Reference Möller2009). They are essential for assessing treatment outcomes and evaluating psychiatric service systems from both objective and subjective perspectives (Kilian et al., Reference Kilian, Bernert, Matschinger, Mory, Roick and Angermeyer2001). In addition to standardised tools, researchers often used specialised measures tailored to capture specific aspects of mental health. For example, the Rosenberg Self-Esteem Scale (Rosenberg, Reference Rosenberg1965) focuses on self-esteem, while the Social Skills Group Assessment Questionnaire (Goldstein and Pollock, Reference Goldstein and Pollock1988) assesses social functioning and interpersonal skills. Measures like the Children’s Depression Inventory (Kovacs, Reference Kovacs1978) and the Reynolds Adolescent Depression Scale (Reynolds, Reference Reynolds2010) cater to developmental differences in symptom expression among children and adolescents.

Furthermore, culturally adapted instruments in Farsi and Krio (e.g., Betancourt et al., Reference Betancourt, McBain, Newnham, Akinsulure-Smith, Brennan, Weisz and Hansen2014; Newnham et al., Reference Newnham, McBain, Hann, Akinsulure-Smith, Weisz, Lilienthal, Hansen and Betancourt2015; Jani et al., Reference Jani, Vu, Kay, Habtamu and Kalibala2016; Moeini et al., Reference Moeini, Bashirian, Soltanian, Ghaleiha and Taheri2019) ensured that assessments are sensitive and relevant to diverse populations (Van Voorhees et al., Reference Van Voorhees, Fogel, Reinecke, Gladstone, Stuart, Gollan, Bradford, Domanico, Fagan, Ross, Larson, Watson, Paunesku, Melkonian, Kuwabara, Holper, Shank, Saner, Butler and Chandler2009; Moeini et al., Reference Moeini, Bashirian, Soltanian, Ghaleiha and Taheri2019). While there were only four studies within this current review exploring this concept of cultural adaptation while conducting psychological measurements, these context-specific assessments further highlight the rising need for validity and reliability of mental health measures, in order to capture unique experiences across diverse groups (Kaiser et al., Reference Kaiser, Ticao, Anoje, Minto, Boglosa and Kohrt2019). This has the potential to facilitate a nuanced understanding of specific domains of mental health, allowing researchers to explore intricate relationships and processes. Recognising the importance of sensitivity to cultural differences and the subjective nature of mental and behavioural disorders (Brugha & Meltzer, Reference Brugha and Meltzer2017) could open up a new avenue to the path of a comprehensive approach that aligns with the broader concept of mental health, encompassing emotional, psychological and social well-being (Doré & Caron, Reference Doré and Caron2017).

Limitations and future directions

While this systematic review provides valuable insights into the current landscape of youth mental health interventions, four key limitations must be acknowledged. First, the inclusion and exclusion criteria, such as the 15 to 24 age range and the exclusion of comorbid conditions, might have inadvertently omitted relevant studies, thereby narrowing the scope and limiting applicability to more complex youth populations. Second, the use of a four-tier classification system (General Youth, Vulnerable Youth, Youth at Risk and Youth with Diagnosis) enabled structured synthesis but might oversimplify the fluid and overlapping nature of youth mental health experiences. Although each study was categorised into only one group for analytic clarity, conceptual overlap among categories is likely. Third, considerable heterogeneity was observed across the included studies, particularly in intervention types, settings and outcome measures. While such variation reflects the diversity of youth mental health needs and service contexts, it might also limit the comparability of findings and the strength of conclusions that could be drawn. Finally, the narrative synthesis approach allowed for the integration of diverse study designs and contextual factors, offering a comprehensive overview of intervention trends and implementation contexts; however, this method did not support causal inference or direct comparisons of intervention effectiveness.

Given these limitations, the findings of this review should be interpreted within the context of its methodological scope, offering a foundation for future targeted research. Future research should adopt a more inclusive and context-sensitive approach to study selection and synthesis, including greater attention to comorbidities and age overlaps. Longitudinal and comparative studies are needed to assess the suitability and relative effectiveness of different intervention strategies over time and across youth subgroups. Efforts should also focus on examining the heterogeneity identified in this review, particularly in intervention types, settings and outcome measures, and assessing their relevance to specific implementation contexts. Transparent reporting of cultural adaptations and implementation processes is essential to distinguish meaningful contextual tailoring from methodological variation. Lastly, expanding research in underrepresented regions and populations, and exploring multi-category or dimensional classification systems to better capture the complexity of youth mental health, will be critical for advancing equity, relevance and impact in future intervention research.