A Systematic Review of Mindfulness-Based School Interventions with Early Adolescents

Significant biological, neurodevelopmental, social and psychological changes occur during the early adolescence developmental stage, the period between 11 and 14 years of age (Patton et al. 2016). Important health and social problems either start or peak from early adolescence through to young adulthood (World Health Organisation [WHO] 2017a). Mental health in particular is a concern, with 20% of young people experiencing a mental health problem within any given year (WHO 2017b). The perceived stigma in accessing mental health services (Lawrence et al. 2015) and the shortage of mental-health professionals (Patel et al. 2007) are known to make it difficult for many young people with mental health issues to receive adequate support. Left untreated, mental health issues can have a significant and detrimental effect on students’ well-being, functioning and development (McGorry et al. 2014). It is therefore important that early intervention and prevention strategies are developed for this age group.

Schools are a universal access point to deliver services promoting health and well-being to early adolescents (WHO 2017b). School-based well-being programs also provide effective support for students in minimising mental health risk (Dray et al. 2014). Mindfulness-based interventions (MBIs) are one such school-based well-being program for promoting the health and well-being of school-aged students. While school-based MBIs differ in program content and outcomes measured, preliminary evidence supports their benefit for the well-being of students (Weare 2013). Reported benefits include an increase in positive affect measures, such as a person’s sense of happiness and optimism post-intervention (Sampaio de Carvalho et al. 2017), as well as reductions in negative affect measures, such as feelings of being afraid or worried (Sibinga et al. 2016).

While MBIs have been delivered to students across diverse age groups, ranging from ages 5 to 19 years of age, only recently have studies suggested that MBIs may be more effective in some developmental stages than in others (e.g. Atkinson and Wade 2015). Carsley et al. (2017) conducted a meta-analysis which found that studies conducted in late adolescence (ages 15–18 years) (n = 7; Hedges’ g = 0.28, 95% CI [.17, .39], p < .001) had higher pre-post effects on mental health and well-being outcomes compared to studies with middle childhood students (ages 6–10 years) (n = 6; Hedges’ g = 0.20, 95% CI [.03, .37]). Their review also found no significant pre-post effects reported in mental health and well-being outcomes in early adolescent students (ages 11–14 years) (n = 6; Hedges’ g = 0.11, p = .213). Differences in the development of self-concept and neurocognitive maturity may influence MBI effects. Differences in self-concept between preadolescent (grades 4–5) and early adolescent students (grades 6–7) may lead to differences in self-awareness (Schonert-Reichl and Lawlor 2010), a key element of mindfulness. Neurocognitive maturity may also influence the effects of MBIs across different age groups (Johnson et al. 2017). Given the developmental significance of early adolescence and the proven effectiveness of school-based well-being programs for minimising mental health risk, it is important to closely examine school-based MBI studies targeted for early adolescents.

There have been several systematic reviews on MBI effects with adolescents, but they included studies implementing MBIs in a broad range of settings (e.g. Klingbeil et al. 2017; Meiklejohn et al. 2012). To date, there have been three reviews conducted exclusively on MBIs within the school environment (i.e. Carsley et al. 2017; Felver et al. 2016; Zenner et al. 2014). In the review conducted by Zenner et al. (2014) significant positive effects were documented in the group design studies, such as improvements in cognitive performance and resilience. Similarly, the review by Felver et al. (2016) reported MBI intervention outcomes, including a reduction in behavioural problems, anxiety, depression, affective disturbances and suicidal ideation, and an increase in executive functioning. Carsley et al. (2017) review also reported small to moderate significant effects from pre-post MBI compared to control groups. Additionally, moderator analysis reported effects from: program facilitator (e.g. teacher or external facilitator) (Carsley et al. 2017), developmental period of student (Carsley et al. 2017) and intervention dosage (e.g. total minutes of mindfulness practice) (Zenner et al. 2014).

All the reviews, however, show methodological and empirical limitations. The diverse range of student ages included in the reviews (e.g. 5 to 19 years) makes it difficult to guide future implementation of MBIs with a specific age group of students or within a particular school environment (e.g. middle school). Given the reports of differences in age effect, a new review is needed to provide a collective account of all MBIs specifically with early adolescent school-aged students and to address some of the methodological and empirical limitations surrounding past MBI school reviews. In addition, Felver et al.’s (2016) review included all MBI studies delivered within a school setting, including those with a ‘general population’ of students, as well as programs specifically delivered to a cohort of students recruited because of behavioural difficulties (Singh et al. 2007), health problems (Lagor et al. 2013) or learning difficulties (Beauchemin et al. 2008). This diversity in student characteristics makes it difficult to determine whether effects reported are specific to a particular cohort of students or to a wider population, which is important given the different effect sizes reported across separate settings with youth (Klingbeil et al. 2017). The lack of quality critique employed by the authors to examine the rigour of each study also warrants a mention. A quality analysis critique can be used in evidence-based practice to inform judgements on the quality of a paper (Hwang et al. 2017). Empirical-based evidence is derived from direct and indirect observations that are provided by both quantitative and qualitative design experiments (Goodwin and Goodwin 2016). The exclusion of qualitative data in these reviews therefore omits valuable information that could be used to better inform implementation of MBIs.

This systematic review aims to provide a collective account of school-based MBIs conducted exclusively with early adolescent school-aged students while addressing the identified methodological and empirical limitations of current reviews on school-based MBIs. In doing so, the review is expected to provide practical and educational implications for practitioners working with early adolescent school-aged students in accordance with the concept of evidence-based practice (EBP). EBP in education is an approach adopted to ensure all aspects of education, from classroom practice to policy making, are based on significant and reliable evidence (Hempenstall 2006). EBP includes professional judgement, client values and best available evidence derived from significant and reliable studies, including qualitative and quantitative research (Slocum et al. 2012). Best evidence requires the most relevant evidence be examined to ensure the highest degree of certainty in decision making (Spencer et al. 2012). Within the field of education, the use of quality criteria can provide a consistent and explicit overview of a study design to guide EBP (Phillips et al. 2016). A source for evidence-based education includes the “What Works Clearinghouse” (Version 4.0) operated by the US Department of Education’s Centre for Education Evaluation and Regional Assistance (NCEE), which evaluates educational programs by both evidence and effectiveness (What Works Clearinghouse [WWC] 2017). The standards developed by WWC can be used to inform educational professionals about the best available evidence as well as limitations in studies. Systematic reviews are instrumental in implementing EBP by providing a transparent research synthesis to education professionals to enable them to make informed judgements about educational programs (Hwang et al. 2017).

The review aims to advance the current understanding of school-based MBIs for early adolescents by addressing the identified methodological and empirical limitations of current reviews on school-based MBIs. Four research objectives were developed in conducting this review. Firstly, the review aims to examine the reported outcomes of mindfulness-based school programs with early adolescent school students (11–14 years). As part of the review process, we will conduct a quality analysis critique of each reviewed study. Existing quality indicators will be applied to examine the methodological rigour of quantitative and qualitative studies under review and new quality indicators of mixed method studies will be developed and applied in their absence. The inclusion of qualitative data in the review will incorporate the experience of students who learn and practise mindfulness into the review outcomes and generate educational implications for teaching and researching mindfulness with this age group. Finally, the review aims to inform educators of the practical implications of implementing MBIs with early adolescent students in the future.

A methodological quality critique was performed on the 13 papers included in this review, using quality indicators developed for quantitative and qualitative studies by authors (Hwang et al. 2017). In the absence of quality indicators for mixed-method studies, new indicators were developed based on the work of Heyvaert et al. (2013), Pluye et al. (2009) and Hwang et al. (2017). The inclusion of the mixed-method design studies in this review is important given the considerable amount of criticism over methodologies employed in mindfulness research to date (Shankland and Rosset 2017; Tan 2016). Quantitative studies have been criticised as being too restrictive in seeking a specific hypothesis, and in so doing possibly missing other outcomes of an intervention (Bernay et al. 2016). Similarly, qualitative studies have been criticised for the small sample sizes and subjectivity reported in the results (Burke 2010).

The two authors independently reviewed each paper against the identified quality analysis indices, using 17 quantitative design indicators, 21 mixed-method (quantitative focus) design indicators and 19 mixed-method design (qualitative focus) design indicators. The results produced an initial agreement on 224 and disagreement on 17 of 241 quality indicators. Further discussions yielded agreement on the 17 indicators that were under dispute. The six mixed-method papers were examined to identify their weighting towards qualitative or quantitative focus to ensure that quality indicators were employed relevant to the study design. The researchers identified four of the papers as having a stronger quantitative focus and two papers as having a stronger qualitative focus. The appraisal was used to indicate to the reader the quality and relevance of each paper and not to discard any of the papers. The results of the 13 papers are presented in Tables 1, 2 and 3 according to the research design of each paper with the quality indicators for each paper referenced as a footnote under each table.

The findings of the seven quantitative studies and the quantitative findings of the four mixed-method (quantitative focus) studies were extracted to examine research methods and intervention effects of the MBI for the health and well-being of early adolescent students. The qualitative findings of the six mixed-method studies (both qualitative and quantitative foci) were extracted and inserted into Nvivo 11, a qualitative data analysis software program, for secondary analysis. Analysis began with (1) reading the extracted primary qualitative findings, (2) generating and reviewing initial codes to elucidate students’ experience of learning mindfulness and teachers’ experience of implementing a MBI, (3) searching for overarching themes that allow consistent patterns to emerge and (4) clustering the initial codes according to their relationships with the identified overarching themes (Braun and Clarke 2006).

This review analysed 13 studies of MBIs with early adolescent school students. The review outcomes are presented under the following three categories, research method, mindfulness intervention and practice and experience. The number of quality appraisal indicators met by the seven quantitative studies ranged from 11 (Sibinga et al. 2016) to 15 (Barnes et al. 2004; Johnson et al. 2017) on a total of 17 measures (Table 1). The quality appraisal indicators met by the four mixed-method (quantitative focus) studies ranged from 12 (Viafora et al. 2015) to 17 (Britton et al. 2014) on a total of 21 measures (Table 2), and those by the two mixed-method (qualitative focus) studies were 6 (Arthurson 2015) to 15 (Costello and Lawler 2014) on a total of 19 measures (Table 3). Such variation indicates that some studies had stronger methodological rigour than others.

This systematic review aimed to provide a collective account of school-based MBIs conducted exclusively with early adolescent school-aged students (11–14 years) to develop practical and educational implications for educators working in accordance with the framework of EBP. In doing so, the paper conducted an extensive review of both quantitative and qualitative research to better understand the reported intervention effects, along with the acceptability and feasibility of the program by including students’ and teachers’ voices. The review conducted a quality analysis critique on all papers reviewed to examine the strengths and limitations that currently exist in research in this field, which will be useful for advancing research design and methodologies on school-based MBIs. The methodological limitations in these studies can also be used to determine whether current MBI research with this age group of students meets the criteria guidelines of effective educational programs in EBP. Finally, the review examined what did and did not work in the papers reviewed to provide educational and practical recommendations for practitioners in the field.

All papers under review applied an MBI to enhance the well-being of early adolescent school-aged students. Emotional well-being was the most targeted major intervention outcome. Overall, MBIs for early adolescents appear to be more effective for decreases in negative mental traits (e.g. affective disturbances, anxiety) than increases in positive mental traits (e.g. positive affect, prosocial functioning). Medium to large effect sizes were reported for decreases in negative mental traits, while small effect size was reported for increases in positive mental traits. This would suggest that a MBI program is a suitable well-being preventative program to be used with early adolescent school-aged students. The acceptability and feasibility of the programs reported by both the students and teachers provide further additional support for a MBI being a suitable well-being prevention program with this age group of students.

The qualitative analysis containing the voices of students and teachers who participated in the studies under review provides further support for the delivery of MBI programs in schools. For the majority of students, the practice of mindfulness was enjoyable and this allowed them to see a range of benefits including decreases in anger, stress and anxiety and improvements in concentration, behaviour management and sleep quality. These benefits were noticed not only by the students who practised mindfulness but also the teachers who implemented the interventions. However, a minority of students disliked the practice finding it boring, difficult to practise and even exposing sadness and fatigue. Teachers also observed that a small minority of students had difficulties in taking mindfulness activities seriously. These findings highlight important aspects to be noted for successful implementation of MBIs. Mindfulness practice may be beneficial for a majority of early adolescent students, but not necessarily for all. Practising mindfulness may reveal underlying difficult physical and emotional conditions among practitioners, so implementation of MBIs requires the requisite knowledge and skills to deal with unexpected challenges.

The use of a quality analysis critique is a useful way to examine the methodological rigour in current studies which can guide future research in this field. In examining the 11 quantitative designed studies (including four mixed-method quantitative focus studies), the main limitations surround the lack of reporting on intervention fidelity, validity measures and measurement of intervention effects at appropriate times. Intervention fidelity is particularly important given the diversity that exists between the programs in content, format and duration. It is also important considering more than one facilitator delivered an MBI program in seven of the 13 studies reviewed. Fidelity of mindfulness interventions can be assessed independently by using a developed checklist (Hwang and Kearney 2015), enabling future studies to be more appropriately compared to one another and ensuring greater confidence in consistency of program delivery to all students in a study.

In examining the two mixed-method (qualitative focus) designed studies, the main limitations surround the lack of reporting on trustworthiness and credibility, a reflexivity statement and the absence of a description of the relationship between researcher and participant. The benefit of employing mixed-method design studies can address some of the limitations that currently surround MBI studies with either a quantitative or qualitative design. Mixed-method design can inform future practices and policies in the field by providing components of both investigation and interpretation. The quality of the methodology in these papers therefore needs to ensure it is rigorous to provide best evidence for MBI research. Whilst the papers examined in this review had a clearly identified quantitative or qualitative focus, it is important that these papers still ensure that they provide appropriate analysis and information on both components of the study. Overall, information on the analysis of both designs was lacking in three of the six mixed-method papers.

These identified limitations are useful for the evaluation of the 13 studies in determining whether they collectively meet criteria required for an intervention in order to be considered as having positive effects within the context of EBP. As a source for EBP in education, the WWC (2017) specified three criteria; (1) statistically significant positive outcomes reported for two or more studies, with effect sizes greater than 0.25; (2) at least one of the studies uses randomised controlled trials, showing comparability of participants between intervention and control groups and low attrition; and (3) none of the studies shows statistically significant or considerably important negative effects (Hwang et al. 2017). The current systematic review demonstrates that these three criteria are not yet fully met.

Of the 13 studies reviewed, seven employed randomised trials, reported attrition and established comparability between intervention and control groups (Barnes et al. 2004; Britton et al. 2014; Johnson et al. 2016, 2017; Quach et al. 2016; Sibinga et al. 2013, 2016). Although overall intervention effects were positive, some effect sizes were lower than 0.25. In addition, two studies (Johnson et al. 2016, 2017) failed to find positive effects and one of them (Johnson et al. 2016) reported negative effects as demonstrated by higher anxiety found for males in the intervention group at follow-up. As these two studies were conducted by the same group of researchers, the intervention content and conditions shared by these studies may have contributed to these unexpected findings. For example, both studies relied solely on self-reported measures to evaluate intervention effects, employed the same facilitator to deliver the program and reported lack of practice engagement in students.

Qualitative analysis findings highlighted the importance of ongoing engagement with practice which may indicate a condition conductive to successful delivery of MBIs. Alternatively, these findings may suggest that an MBI may not have positive effects on the well-being of early adolescent school students. This possibility is concerning given the time and resources of implementation. Since other studies reported positive effects of MBIs for the mental and physical well-being of early adolescent students and the students’ own experiences support such claims, further investigation is required before drawing any conclusion about whether MBIs meet the criteria of EBP as an effective educational program.

The other key purpose of this study was to identify the best way to introduce and practise mindfulness with this age group. To fulfil this purpose, it is important to look at what has worked in the studies reviewed. The programs that were delivered to students all varied extensively in content, format, structure and duration. While the core elements of mindfulness, ‘present moment awareness’ and ‘breathing awareness’ were highlighted in the majority of the studies, that was where the similarities ended. Only four studies reviewed had curriculum manuals, and one study reported a descriptive approach to the program allowing the teacher to tailor the program to their class. The two studies that reported no positive improvement from pre- to post-test measure were the only two studies in this review to employ the Mindfulness in Schools Project ‘.b’ curriculum. While this program has reported positive effects with older aged adolescent students (Kuyken et al. 2013), the content and format of different mindfulness programs may have varying optimal age ranges for implementation in schools (Johnson et al. 2017). This needs to be investigated in future studies.

The duration of the programs reviewed varied considerably from the number of minutes practised at each session to the frequency of sessions. Positive improvements were reported across programs irrespective of the duration of the lesson or its frequency. This is in line with a recent review of MBIs with adolescents across different settings (Klingbeil et al. 2017), which indicated that intervention dosage may not be nearly as significant as other elements of MBI intervention, such as program facilitators.

The program facilitators in this review were evenly divided between teaching staff and external facilitators. They all had varying degrees of personal practice, training and experience, with external facilitators generally providing more information on their credentials. The only two studies that reported no positive effect were delivered by the same external facilitator (Johnson et al. 2016, 2017). This may indicate that different facilitators will bring their own nuances and personality to the delivery of the program and that this needs to be considered in examining an instructor’s suitability for implementing a MBI. In Carsley et al. (2017) review, they reported significant effects on mindfulness post-intervention only in studies with an external facilitator, and significant effects on mental health outcomes post-intervention in studies only with trained teacher facilitators. This again suggests that the facilitator plays an important role in program implementation as they may be more knowledgeable in the material being delivered in some subjects than others. Secondary analysis of qualitative findings stresses student engagement as a key to positive intervention outcomes, which presents a more complicated picture concerning the requirements of facilitators than this single binary approach (i.e. external facilitator vs teacher facilitator).

Teachers nominated student disengagement as a barrier to successful delivery of a MBI. A small group of students found it difficult to engage with practice for a variety of reasons. Of these, difficulties reported by students in dealing with the wandering mind and underlying physical and emotional conditions emphasise the importance of facilitators’ knowledge and skills. They need to be equipped with not only the knowledge and skills of teaching mindfulness (e.g. being able to explain that a wandering mind is part of mindfulness practice) but also those of students (e.g. student health and educational backgrounds) in order to adequately address any challenges that can lead to disengagement. In addition, teachers’ ability to embody mindfulness was noted as an enabler, along with support from school administrators and parents, a relaxing environment and students’ willingness to learn. These findings suggest that it is important to create conditions that are conducive to successful implementation of MBIs.

The search strategy employed in this review limited the search with the inclusion criteria of English language and peer-reviewed articles only. Therefore, whilst an extensive search was conducted, it is incorrect to claim that the review is exhaustive given the papers excluded that were written in other languages or articles published in other formats (e.g. unpublished theses). The review included all papers that meet the search strategy criteria regardless of the quality of the paper. As discussed in the results section of this review, the quality appraisal index that was assigned to each paper varied significantly between the studies. Some papers demonstrated stronger research rigour than others. However, the review included the findings of all of the papers under review to provide a comprehensive picture of MBIs for early adolescents. Interpretation of the systematic review results therefore requires caution.

Despite these limitations, this systematic review presents a synthesis of MBIs that have been delivered specifically to early adolescent school-aged students. Findings of the review demonstrate the potential of MBIs as a tool for enhancing the well-being of early adolescents. They also highlight practical implications for educational professionals who aim to enhance the quality of life of early adolescents through the implementation of MBIs at school. Identified methodological limitations of the existing MBI studies call for future research and this will provide necessary information regarding whether MBIs are an effective educational program for early adolescents within the framework of EBP.