Intervention Fidelity in Mindfulness-Based Research and Practice: Overview of the Special Collection and State of the Field

Abstract

Mindfulness-Based Programs (MBPs) are increasingly being introduced into a range of mainstream contexts such as health care, education, criminal justice, workplace, community settings, and leadership. MBPs are available in some form in every continent, and in some countries have become established as part of standard care in publicly funded health care systems and routinely embedded into school curriculums. Efficacy and effectiveness research demonstrate gold standard evidence in some areas and significant promise in others, and cost-effectiveness evaluations are encouraging. However, understanding how to assess and ensure the fidelity of program delivery is at a germinal stage. Without clear, robust, and coordinated approaches to this issue within the international field, ongoing dissemination and implementation could be compromised by variable standards of teaching skill and adherence to program form. These issues are relevant in practice contexts where approaches to teacher training internationally need to adhere to commonly held principles, understandings, and standards. They are also relevant in research contexts where assessment of intervention fidelity needs to ensure that the variable of the teaching process and curriculum is addressed in the research design and conveyed to readers in the presentation of data. In this special collection, we bring together articles reporting on research and practice on MBP intervention fidelity in teacher training, implementation, and development of governance for the emerging mindfulness-based field. This paper introduces the special collection, summarizes the papers included, and offers an analysis of where we are now as a field and of what steps are needed to further our understanding on these issues. We hope to share current best practice and catalyse new understandings and directions of inquiry that will provide a compass to navigate forward and benefit this promising field.

Introduction

The Mindfulness-Based Program (MBP) field has established itself as a major contributor to human flourishing and wellbeing. Simultaneously, it is also clear that the field is at an emergent stage and that some key vulnerabilities could undermine the promise and potential of MBPs. One vulnerability is the rapid expansion in MBP research and practice with relatively little attention being paid to assuring the quality of the intervention. This is a challenging issue for any intervention in the transition from research to practice. In the MBP context there are particularities in this arena that need attention. MBPs place a substantial emphasis on MBP teacher qualities because of the distinct pedagogical approach, which rests on the teacher’s capacity to convey mindfulness through both how they teach and what they teach.^1,2 Thus, the skill of the teacher in embodying mindfulness (as compared to conveying concepts cognitively through verbal instruction) is hypothesised to be central to the effectiveness of MBPs. Despite the central importance given to this issue in MBP teacher training contexts, teacher factors have received little study in research contexts.

One issue underlying both the importance and vulnerability of the fidelity (also termed intervention integrity) of MBP delivery is the rising popularisation of mindfulness. There are several markers that illustrate the global expansion of research and practice in the MBP fields. ³ Mindfulness-based teacher training and courses for local populations are now available on all 6 of the world’s populated continents. Mindfulness-Based Cognitive Therapy (MBCT) is embedded into clinical guidelines for delivery in health services in the UK, ⁴ USA, ⁵ Canada, ⁶ Belgium, Australia, ⁷ New Zealand, ⁷ and Brazil. ⁸ In England, MBCT is mandated for delivery in the National Health Service, with central government funding supporting the training of therapists to build capacity. ⁹ The UK parliament has an All-Party Parliamentary Group investigating the policy implications of mindfulness, and has published a report laying out the relevance of mindfulness to public life. ¹⁰ MBPs are now offered to a wide range of populations, in a breadth of mainstream settings including education, health care, the justice system, workplaces and within the private sector. More generally, the popularisation of mindfulness can be seen by publications such as The New York Times ¹¹ and The Sunday Times, ¹² which have included articles offering instructions for mindfulness meditation; and smartphone apps on mindfulness such as Headspace, which has been downloaded over 80 million times, ¹³ and Calm, which has been downloaded over 133 million times by 2023. ¹⁴ This growth in the popularity of MBPs has been encouraged, in part, by the growing body of scientific evidence for benefits of these programs. The number of articles with mindfulness in the title has expanded year on year in the last decade. A search of the Cochrane Central Register of Controlled Trials on 16^th January 2023 with mindfulness in title, abstract and keywords revealed that since 2010 to the present there were 11 764 registered trials, whereas in the previous decade there were just 477. While popular views of the benefits of MBPs sometimes outstrips the actual evidence, we now have well-designed trials that have shown benefits in areas like depression, ¹⁵ anxiety, ¹⁶ pain, ¹⁷ and workplace well-being. ¹⁸ There are tensions between the necessarily slow and steady development of the science, and this rapidly expanding public interest. As mindfulness becomes more embedded in mainstream institutions and society, the potential impact grows, but also the potential to lose the original intentionality and integrity of programs on which the evidence for benefit is based.

The MBP field represents a convergence and integration of different epistemologies. ¹⁹ On the one hand, MBP research and practice is embedded in a commitment to contemporary Western empirical methods, and to implementation within mainstream institutions. ¹⁹ Simultaneously, on the other hand, the approach is embedded in a paradigm of non-fixing, non-duality, and non-goal orientation. In the opening paper to a previous special issue on the MBP field in Contemporary Buddhism just over a decade ago, Williams and Kabat-Zinn expressed how vital it is that the integrity of each influence is deeply respected: “The world can only benefit from such a convergence and intermixing of streams, as long as the highest standards of rigor and empiricism native to each stream are respected and followed.” ²⁰ This integration of different ways of knowing brings a range of creative tensions and challenges. The themes of teacher skill, intervention fidelity and program fidelity are at the core of these challenges. MBP teachers sit at the fulcrum point, bridging the paradigms of the MBP teaching process with those of the mainstream contexts they operate within.

The transformational potential of MBPs is hypothesised to rest upon their paradigm or mode of mind shifting potential – ie, that participants may discover radically new ways to orient around existing life issues and challenges. A vulnerability of the MBP field may be that in the process of finding their place in the mainstream, compromises are made in areas such as teacher training or program delivery length, which could mean that the approach is no longer fundamentally based within a “being mode of mind orientation” and so the full depth and potential of mindfulness training may be missing.^21,22 The critical mindfulness ingredient of enabling the individual to make a paradigmatic shift in orientation and perception towards life challenges, which depends in part on the ability of a skilled MBP teacher to facilitate such shifts through guiding practices thought to mediate some of the benefits of mindfulness programs, such as being able to step back from observe and one’s thoughts (de-centering), ²³ may be missing. It may still be offering benefit in other ways, but it is no longer mindfulness-based – it is now based on other mechanisms of change.

In the early stages of investigating MBPs, the science and the delivery of the programs were kept quite separate. The primary question being asked during this initial research, corresponding to Stage 1 (Intervention Generation/Refinement) and Stage 2 (Efficacy in the Research Clinic) in the NIH Stage Model for behavioral therapy research, ²⁴ was focused on intervention refinement and establishment of efficacy: “Does this approach work? Do these programs enable shifts in key markers of health and wellbeing?” The answers were highly encouraging. However, as the research process has advanced, it has become increasingly important to bring the magnifying glass closer to the very process enabling these beneficial outcomes – the MBP teaching process and its fidelity. These issues are particularly important in Stage III (Efficacy in Community Clinics), Stage IV (Effectiveness), and Stage V (Implementation and Dissemination) research on behavioral interventions. ²⁴ Questions we need to ask include: How do we recognise skilful MBP teaching? Is it possible to assess the skills involved in MBP teaching in a consistent, quantifiable, and valid way? How do we assess whether an MBP is delivered with fidelity? Do higher levels of competence and fidelity correlate with better participant outcomes? Which aspects of the program and/or the skill of the teacher are most critical in enabling participants to do well in an MBP? It is also important to recognise that positive outcomes do not only rely on the teacher – we need to develop our understanding of the range of variable factors influencing learning and engagement, including those within participants themselves.

In the early days of delivery and research, MBPs that were the subject of research were delivered by the developers or by carefully selected close colleagues, so it was clear that they were being delivered as was intended.^25,26 We are in a different phase now. Delivery is being undertaken by generations of teachers a long way removed from the originators of the founding MBP curriculums. This widening of research and implementation activity is an advance. It comes, however, with the challenge that with increasing numbers of teachers, ensuring intervention fidelity becomes both more complex and more vital in research and in practice settings. A 2015 mapping of the MBP evidence noted the “lack of attention to developing formal measures, methods, and standards for determining instructor quality” as one of the critical issues in advancing MBP research and implementation. ²⁷

MBPs have tended to be developed and disseminated in ways that deliberately facilitate global uptake. For example, the two most empirically supported and widely implemented MBPs, Mindfulness-Based Stress Reduction (MBSR) and MBCT, are not controlled by a particular institution or teacher. We believe this “open system” is a critical reason why the field has flourished and had the impacts we are observing. But as the field has grown, this has led to other challenges about how to ensure program quality while preserving this “open architecture.”

Our aim in bringing together this collection has been to:

(1) Raise the issue of intervention fidelity up the agenda and awareness of MBP researchers and practitioners;

(2) Catalyse conversation across the research/practice boundary;

(3) Capture the current state of the art in the field in terms of research, practice and thinking on intervention fidelity, particularly as it relates to teacher training and skill;

(4) Encourage the integration of methods for assessing and reporting on intervention fidelity into MBP research;

(5) Provide perspectives on the integration of methods to assess the skill of MBP teachers in teacher training and dissemination of MBPs.

(6) Open inquiry on the alignment between teaching fidelity and the work of ensuring that the forward development of the field is embedded in the ethics of justice, equity, and inclusion.

(7) Offer an analysis of the key challenges and potential ways of addressing these going forward.

Many of the papers in this collection report on the Mindfulness-Based Interventions: Teaching Assessment Criteria (MBI:TAC). ²⁸ While we view the MBI:TAC as an important tool for intervention fidelity assessment in teacher training and in research, we want to acknowledge that this focus comes in part from the perspectives and backgrounds of the two of us as editors of this special collection. One of us (RC) led the development of the MBI:TAC, and both of us have been involved in research on it. Despite this focus, we want to emphasize our hope that this special collection broadens rather than narrows the discussion of methods for assessing and understanding intervention fidelity in MBPs. We are delighted that several papers report on the development of new fidelity tools and methodologies, and examine innovative developments based on the “DNA” of the MBI:TAC.

Introducing and Commenting on the Papers

We organise the papers within the special collection ²⁹ around 4 overarching themes/sections. First, to reflect the broad issues and themes that are at the heart of implementation fidelity questions we open with 4 papers that analyse some of the questions, challenges, and tensions inherent in this research-practice interface, in a section titled: “Big picture issues related to intervention integrity.” Second, the issue presents 4 papers in a section titled: “Implementing the MBI:TAC in practice,” that offer perspectives on implementation themes, challenges and dilemmas when using the MBI:TAC both as a tool to formally assess teaching skill and program adherence, and as an informal support for reflective practice and skill development. Third, in a section titled: “Research on MBP fidelity tools” we present 7 research papers examining the implementation of the MBI:TAC and other fidelity tools. The fourth and final section of the special collection titled: Implementation and Governance, includes 4 papers reporting on implementation and governance themes related to teacher qualifications and intervention fidelity in MBP practice.

We now provide an overview and key “take home” messages of each paper within this structure.

Big Picture Questions Related to Intervention Fidelity

In the first paper, Monteiro ³⁰ makes a plea for bringing a greater level of relational process into the assessment of intervention fidelity. The method currently used to assess fidelity is generally via observation of the teacher by an assessor using a set of criteria (as is used by the MBI:TAC, for example). Monteiro encourages the development of methods that explicitly include the experiences of participants by asking them for their perceptions on the competence of their teacher. The potential benefits of including participant experience as a component of the assessment may be particularly significant in enabling future delivery to be more tuned to reducing any potential for harm and increasing alignment towards participants’ cultural and faith values. Monteiro’s perspective is currently being examined in a development led by researchers at Oxford University. They are carefully adapting the MBI:TAC with input from international collaborators with the aim of developing a tool that can enable MBP participants to validly assess the skill of their MBP teacher. This new tool is provisionally entitled the Mindfulness-Based Interventions: Participant Assessment Tool (MBI:PAT). Future research on intervention fidelity could involve triangulating data from different perspectives: ie, teaching skills assessment conducted via observational assessment with a tool such as the MBI:TAC, participant led assessment with a tool such as the MBI:PAT, and teacher self-assessment with a tool such as the newly adapted version of the MBI:TAC specifically designed for personal reflective practice (the Mindfulness-Based Interventions: Teaching and Learning Companion) ³¹ (see below for more discussion on this). Data from these 3 perspectives offers the potential to advance understanding about optimal methods of assessment going forward; and as Monteiro points out, it may be that the information gathered from different perspectives offers important nuances of understanding that will inform future program delivery.

In the second paper in this section, Loucks et al ³² pick up a key challenge related to program fidelity: how to adapt MBPs for new populations and contexts in ways that are embedded in good practice. The paper is presented as a sister paper to the earlier “Warp and Weft” paper, which laid out a definition of the ingredients of any program that is titled “mindfulness-based,” using a framework of laying out the essential (warp) elements, and variant (weft) elements. ² Loucks et al also use this framework to structure and guide the necessary thinking about how to adapt an MBP. There is often a forward momentum and enthusiasm to develop new programs. The paper urges prospective adapters to first pause and check – why is an adaptation needed? Does the theoretical premise of current MBPs apply to this potential new population? Do the benefits of the proposed adaptation outweigh the costs? Is there already an evidenced-based approach to address the target problem? If after this analysis there is a strong case for pursuing the development, the paper offers an important guide on how to approach the task, and who to involve in the team process. This paper links well with Boxtel et al.’s paper ³³ in the fourth section of this collection in which they report on work to create a governance system for acknowledging new program developments within the context of MBP national associations (see below).

The third paper in this section, by Crane, et al, picks up the theme of adaptation and fidelity and its relationship with equity, inclusion, and diversity both in the context of the MBSR program, and more widely across the MBP field. ³⁴ The paper is written by a US-UK collaboration of MBP trainers from racially diverse backgrounds who have been involved in teacher training initiatives with and for racially minoritized individuals. It examines a tension at the heart of fidelity: how far can one adapt before the need to retitle the program as an adaptation? The writers propose that a defining feature of MBP teaching is responsivity to this teaching moment, group, culture, and context. So, if the essential (warp) elements are all in place (and these are laid out in the paper in the context of the MBSR program), and the tailoring is in the service of enabling this group to connect more fully with the teaching themes, then fidelity is, by definition, high. The paper also makes a call to expand the lens beyond the cultivation of personal wellbeing, into teaching MBPs in ways that communicate the deep interconnection between personal and collective wellbeing. The paper includes a linked resource for MBP teachers to support the learning needed to enable this shift in emphasis.

In the fourth paper in this section, Broderick and Schussler, who have a long engagement with developing and researching MBPs for school-based settings, take a meta-perspective on the development of fidelity assessment for MBPs designed for school delivery. ³⁵ They identify and examine 4 challenges in ensuring fidelity in delivery of MBPs in schools: the lack of consensus in the field on the definition and intention of mindfulness training in this context; balancing the need for flexibility with the need for adherence; understanding the role of the level of the MBP teacher’s mindfulness experience in determining the quality of the delivery; and the importance of delineating the distinctive features of the pedagogy of the delivery.

In the fifth paper in this section, C. Crane led a team on a project embedded into the MYRIAD trial investigating mindfulness training for students in schools. ⁴² This study set out to investigate the effectiveness of 4 different models of training school teachers to deliver the mindfulness program in school settings. It compared an existing, intensive 8-session expert-led course in mindfulness combined with a 4-day training program in the school mindfulness proram for student with 3 progressively less intensive training processes. The goal was to test whether less intensive training processes that were more easily scalable would provide adequate competency to teach the mindfulness program. It used a modified version of the MBI:TAC to assess competency in teaching the MBP. This research addresses some key questions in disseminating MBPs. It suggests that abbreviating teacher training to increase scalability may be unwise, as even the most intensive training model brought less than a third of teachers who began the training to an advanced beginner level of competency or better, and competency levels were lower with the less intensive training models. The study also suggested that after training many new teachers require additional support to support competence, underlining the need for ongoing supervision. ⁴³ Further implications of these findings are picked up in our discussion section.

The sixth paper in this section by Serpa et al, ⁴⁴ examines the integration of fidelity assessment in a large-scale implementation process of MBPs for veterans. As they underline, it is vital to ensure teacher quality when implementation is happening at scale because the impact is far reaching. The challenge though is what fidelity assessment methods are available that are cost-effective and pragmatic. They therefore conducted research examining the reliability of self-assessments conducted by the MBP teachers themselves, compared with independent assessments conducted by trained MBI:TAC assessors. The independent assessors gave consistently lower scores than the teachers gave themselves, with the less competent MBP teachers being particularly prone to over-rate their competence, indicating blind spots in self-knowledge. However, despite the caution provided by these initial results, the authors encourage the field to continue to pursue the use of self-assessment, with the suggestion that integrating training on how to do this reliably and validly into MBP teacher training may help increase the accuracy of the process, and simultaneously is a useful method for the teachers to build reflective development. The MBI:TLC (reported above) ³¹ is now freely available and is explicitly designed to support compassionate self-reflection on strengths and learning needs as an MBP teacher.

The seventh and final paper in this section by Boxtel, Sansom, and Kersemaekers examines how the tensions around MBP adaptation highlighted in Loucks et al ³² interface with the work of professional associations that are designing and implementing governance structures for the MBP field. ³³ The Dutch and UK professional bodies worked together to design a systematic process for reviewing the suitability for newly adapted MBPs to be publicly recognized and listed by the associations. They describe in the paper this “work in progress” development, the questions, and dilemmas that the process revealed, and how they responded to these. Central to their inquiry is navigating the appropriate balance between supporting innovation and maintaining fidelity. This is an important example of how theoretical and empirical developments can inform important steps to build professional structures and governance that support fidelity.

Implementing the MBI:TAC in Practice Settings

The 4 papers in this section offer guidance on implementing the MBI:TAC in practice. First, Crane, et al (unless otherwise noted, Crane refers to R. Crane) lay out current thinking in a viewpoint paper on the systems and governance that are needed to conduct fidelity assessments within both practice and research contexts in ways that are embedded in good practice and an ethic of care. ³⁶ The process of conducting assessments of teaching quality inherently makes the teacher vulnerable. It is critical to consider how the process is held and conducted. Second, Griffith et al ³⁷ examine the contribution that the MBI:TAC can make when it is introduced as an informal reflective tool within teacher training pathways. Although the MBI:TAC was originally developed as an assessment tool, in practice it offers a helpful map of the territory of teaching skill and as such has a range of informal applications. The paper articulates a range of creative ways that the framework can be used in MBP teacher training processes. Third, Evans, Griffith, Crane and Sansom present various practical ways of introducing the tool within MBP supervisory and peer reflection contexts. ³⁸ Finally, Griffith and Crane ³¹ offer a brief viewpoint paper introducing the recent adaptation of the MBI:TAC - the Mindfulness-Based Interventions: Teaching and Learning Companion (the TLC) – a version of the tool explicitly designed to support MBP teacher reflective skill development. Given the range of creative uses of the tool in practice, there was call for a version that took out the focus on assessment and foregrounded the ways the tool supports reflective practice. All these papers report on both the potential benefit that the tool brings, and conversely the potential risks of harm if it is used in unskilful ways.

Research on MBP Fidelity Tools

In this third section of the collection, we present 7 papers reporting research on MBP fidelity tools. First, Hecht et al ³⁹ report the primary outcomes of the “Predictors of Outcomes in MBSR Participants from Teacher Factors” (PROMPT) study. This study is the first research trial in the MBP field explicitly designed to ask questions related to the impact of teacher skill. The PROMPT study set out to answer some key uncertainties related to the performance of the MBI:TAC to determine whether it is ready for broader scale application in MBP research and practice. The research focused on further evaluating the inter-rater reliability of the tool, andaimed to begin addressing a question that goes to the heart of why the investigation of intervention fidelity matters: do MBI:TAC assessment outcomes predict participant outcomes? This was in part a feasibility study to help inform future larger trials. The results found that higher ratings of teacher skill on the MBI:TAC were associated with greater decreases in participant anxiety at the end of MBSR courses. Further work is needed with larger sample sizes to better assess how well the MBI:TAC assessments predict participant benefit.

The second paper by Crane et al reports on outcomes of the MBI:TAC assessor training process implemented within the PROMPT trial. ⁴⁰ The paper reports on the process used, and its effectiveness in bringing trainees up to reliability in using the MBI:TAC tool, and how this learning has been taken forward into practice. Critically, the study found that it was feasible to train an internationally diverse cohort of senior MBSR teachers in the use of the MBI:TAC, and achieve a high level of inter-rater reliability when rating test videos, thus reaching alignment on what skilful MBP teaching looks like.

The third paper in this section by Floyd et al reports on another sub-study within the PROMPT trial that asked the question: do we need to assess teaching via video recordings, or do audio recordings enable accurate assessment? ⁴¹ This is an important practical question because it is often easier and less intrusive to capture audio than video recordings of MBP classes. The findings show that inter-rater reliability using audio recordings was adequate for many research and clinical settings, but that because video recordings offer more information, they are particularly helpful when assessing less experienced teachers. The choice of using audio or video recordings thus depends on the how challenging it is to obtain video recording in a particular setting, how precise the MBI:TAC ratings need to be, the level of experience of the teacher, and the overall purpose of the evaluation. Using video recordings of teaching practice is a particularly valuable part of teacher training both to enable nuanced reflection in the context of formal and informal assessment, and self, peer and trainer reflection and assessment. For researchers who need headline measures of levels of intervention fidelity in their trials and who are using experienced teachers, an audio recording may well be sufficient.

The fourth and fifth papers in this section report the development of and research on other MBP fidelity tools. Greco et al ⁴⁵ report on developing the Concise Fidelity for Mindfulness-Based Interventions (CoFi-MBI) tool. This contains questions about session elements such as meditation guidance and group discussion, and questions about participant engagement. Like the MBI:TAC, it is intended to be used to evaluate recorded MBP sessions. However, it aims to offer less time consuming, simpler fidelity assessments than the MBI:TAC that can be implemented by trained study personnel with less extensive MBP teaching expertise than the MBI:TAC, with the aim of providing a cost-effective fidelity tool that could be more easily used in large clinical trials and dissemation efforts. The kappa statistic for 5 of the 9 overall session domains of the CoFi-MBI were in ranges considered as substantial agreement (0.61 – 0.80) to almost perfect agreement (0.81 to 1.00) ⁴⁶ in this initial study. However, 3 of the 9 domains had kappa statistics below 0.4, generally considered to indicate slight agreement or less, suggesting that further refinement of this tool is needed.

In the seventh paper, Klatt et al ⁴⁷ report on the development of a fidelity assurance system tailored to an adapted MBP for health care professional wellbeing. Interestingly, when the fidelity monitoring system for this program was being developed, the researchers were unaware of the MBI:TAC, but nevertheless included some of the same key domains of the MBI:TAC, such as the ability to facilitate group discussion. In contrast to the MBI:TAC, the fidelity monitoring system they developed involves a more detailed checklist largely focused on specific steps to be implemented in each session that could be rated by a trained observer without substantial expertise in MBPs. Further data are needed to assess inter-rater reliability for this instrument, but the simplified checklist approach may facilitate inter-rater reliability. As noted earlier, this type of detailed checklist approach focuses largely on adherence to program content, and may miss more nuanced aspects of teacher skill that tools like the MBI:TAC are designed to capturein

Implementation and Governance

In this final section of the special collection, we first present two papers reporting on the research of MBCT implementation in the health system in Sweden. The first of these papers (Niemi, Crane, Sinselmeijer and Andermo), ⁴⁸ reports the quantitative aspects of the data, and the second (Andermo, Crane, and Niemi) ⁴⁹ reports the qualitative data. The number of studies on the processes influencing successful MBP implementation globally are small, and these papers offer important perspectives guiding the process. Of course, some of the perspectives are Sweden specific – but the take home message from this is that wherever we are implementing, it is important that we understand and work responsively with the constraints and parameters of the cultural and organisational context within which we are operating, and that implementing new innovations in complex organisational systems takes deliberate and focused effort.

In the third paper in this section, ⁵⁰ Kenny, Luck and Koerbel report on an international collaboration aimed at establishing a network to uphold the fidelity of MBP developments globally, which includes agreeing on a broad set of ethical standards for MBP teachers and trainers, and criteria and standards for teacher trainers and teacher training pathways. This is the first time that an attempt has been made to create organisational and values join-up at an international level. ⁵¹ As well as presenting an important synthesis of understanding from diverse geographical regions on what fidelity looks like in the MBP field, the authors point to the collaborative spirit that informed the process of the engagement. The group created a vessel within which this delicate work can take place respectfully through non-hierarchical dialogue, and closely aligned to the principles of mindfulness practice, rather than linking to business models or competitive market strategies. They also name a series of challenges and tensions integral to the process of simultaneously strengthening MBP fidelity whilst also increasing accessibility, including the need to attune to regional and cultural diversity; affordability; preventing gradual fidelity drift; managing the fidelity of new MBP adaptations; and finding ways to cohere internationally around these challenges in the mindfulness program field.

Discussion

In the following section we step back and consider the papers in this special collection in context of the existing small but significant body of empirical and practice-oriented literature on mindfulness-based program fidelity and teaching competence. Second, we consider potential future research and practice steps for the field.

The Special Collection in Context

Empirical Literature

In our 2018 paper, ⁵² we highlighted a handful of studies that began to build an evidenced-based understanding of teacher training and skill in relationship to participant outcomes in mindfulness-based programs (eg, van Aalderen et al ⁵³ and Ruijgrok-Lupton et al ⁵⁴ ). The C. Crane et al paper on schools within this collection adds to this literature by researching the relative effectiveness of different training models. This study offers a model we encourage others to adopt – ie, embedding practical questions about training and implementation into the design of effectiveness trials. Since our 2018 paper, Bisseling et al conducted a trial that examined 3 factors likely to be related to participant outcome in MBCT programs: group cohesion, therapeutic alliance, and MBCT leader skill/competence (the latter using the MBI:TAC). ⁵⁵ Fascinatingly, the findings revealed that therapeutic alliance and not group cohesion significantly contributed to reduction of psychological distress in MBCT for cancer patients. Therapist competence did not predict post‐treatment psychological distress and did not appear to be a precondition for a good therapeutic alliance and high group cohesion. These kinds of unexpected results underline the importance of continued investigation of teacher and participant factors that influence therapeutic change so that future training and implementation is empirically guided. Bisseling et al acknowledged that their research may have been confounded by the independent raters observing video clips of the teaching potentially having a different perspective on relational skills that the participants themselves. Future research needs to include refining the methods for gathering data on these subtle and nuanced issues.

Again, since our 2018 paper, Ketcher et al conducted a literature review of how fidelity has been reported in the main outcome articles of efficacy trials among adult participants. ⁵⁶ Out of 202 articles included in their review, only 25 described treatment fidelity systems within the research process. Similarly, a recent systematic review by Emerson et al ⁵⁷ aimed to synthesize the literature on the implementation of mindfulness-based interventions in schools. Intervention fidelity was reported in only 45% of the studies. Authors of both papers underline that this lack of detail and consistency in reporting on fidelity, teacher competence and intervention fidelity, across studies limits the advancement of the science. Ketcher et al offer a framework for reporting intervention fidelity the “Treatment Fidelity Tool for MBIs” adapted from the National Institutes of Health Behavior Change Consortium (BCC) guidelines to help researchers monitor and report these methods and measures in a simple and standardized format. In our 2018 paper, ⁵² we presented another framework for reporting intervention fidelity in MBPs - an adaptation of the “Template for Intervention Description and Replication” (TIDieR) from the CONSORT guidelines (Consolidated Standards of Reporting Trials). Both frameworks are rooted within the broader context of how to address and report on fidelity related issues in behavioural research and share many similar elements. Many of the articles in this collection focus on the issue of teacher skill as central component of intervention fidelity in mindfulness-based program delivery. Both the framework in Ketcher et al and our proposed framework for reporting fidelity based on the TIDieR guidelines include additional elements that are less explicitly addressed in this collection, such as describing the rationale and theoretical framework of the intervention, which provides a foundation for assessing which elements of an intervention are conceptually important to have fidelity to. A key point is that researchers need to adopt a structured framework within which to report on intervention fidelity; we refer readers to these earlier articles for further details.

In our 2018 paper we presented a table summarising the current fidelity tools used for adult populations within the MBP field in a table. ⁵² There are also tools that have been developed for MBPs delivered in school settings. For example, Broderick et al ⁵⁸ have developed and published a tool for use within MBP delivery in schools – the “Teaching Mindfulness in Education Observation Scale” (TMEOS) a 28-item tool that integrates qualitative and quantitative aspects of mindfulness instruction. Doyle et al ⁵⁹ report on a tool developed to assess the Cultivating Awareness and Resilience in Education (CARE) MBP (an adaptation of MBSR for school settings). Additionally, as reported within this special collection, Greco et al, Serpa, ⁴⁴ and Klatt et al ⁴⁷ have developed new fidelity tools tailored to MBPs for adult populations. Both examined the implementation challenge of scaling up the delivery of MBPs in workplace settings whilst simultaneously ensuring that the fidelity of the intervention is maintained. In both contexts, the fidelity assessment tool was designed to enable scalability and address two important potential limitations of the MBI:TAC: (i) the MBI:TAC requires substantial assessor time and expertise to perform as currently implemented, and (ii) the MBI:TAC foregrounds teaching skill, and rather than lengthy adherence checklists, it relies on the assessor having depth familiarity of the curriculum they are assessing to enable accurate assessment of adherence to the elements of that particular curriculum. While both of these new tools may provide additional promising tools for assessing MBP fidelity, some of the limitations noted earlier underline that additional work is needed to refine and further validate simplified and more easily disseminated tools.

Practice Oriented Literature

In our 2018 paper, ⁵² we commented on work that is happening regionally across the world to build consensus on governance and standards for good practice and fidelity for the MBP field, which has subsequently continued to develop (eg, British Association of Mindfulness-Based Approaches, ⁶⁰ the European Association of Mindfulness Based Approaches, Mindfulness, ⁶¹ and the Mindfulness Teachers Association of Ireland ⁶² ). The Kenny, Luck and Koerbel, and the Crane, Koerbel and Sansom papers in this collection add to the practice-based literature by reporting on the ongoing process of developing professional governance and frameworks for this emerging field. A key challenge for the field is how to balance the potential tension between implementing systems that support alignment to teaching and training good practice on local, national, and international levels with the open organic architecture of the field. This includes the issue of certification or licensing of MBP teachers as is required in most professional fields. Some countries have organised associations that implement such processes. For example, in the UK the British Association for Mindfulness-Based Approaches runs a listing of teachers meeting agreed upon standards that involves initial screening of credentials and a requirement for evidence of ongoing professional development. ⁶⁰ A similar professional body has also developed in Ireland. ⁶² These national developments require considerable effort and a depth of cooperation between training centres involved.

Literature on Implementation Factors

The implementation of research evidence is complex and involves multiple factors. ⁶³ It is generally under-researched in the mindfulness field, though there are a handful of significant implementation trials (eg, Rycroft-Malone et al), ⁶⁴ and commentary papers urging the field to attend to implementation science and practice in mindfulness-based research (eg, Jennings, ⁶⁵ and Dimidjian and Segal ²⁷ ). The issue of how intervention fidelity can be maintained when programs are implemented outside the original research context in which efficacy was established is an important element in implementation science. ⁶⁶

The C. Crane, et al paper ⁴² in this collection is an important example of the challenges of maintaining intervention fidelity when scaling up a MBP for broad dissemination. This study was embedded into the “My Resilience in Adolescence” (MYRIAD) trial, ⁶⁷ which evaluated the scale up of an existing MBP designed for delivery in schools. ⁶⁸ This curriculum was previously the subject of a controlled trial with 522 pupils, taught by 9 teachers in 9 different schools, which showed a positive impact on pupils’ mental health. ⁶⁹ However, the MYRIAD trial reported a null result: ie, there was no evidence that the mindfulness delivery was superior in effectiveness to usual social-emotional learning lessons delivered in publicly funded schools. MYRIAD was a large (85 schools with over 8000 students) high-quality trial and thus provides strong evidence that the universal roll out of mindfulness training in the form of this curriculum or implementation strategy was not effective. Pragmatically, the level of mindfulness training for the schoolteachers to equip them to deliver the intervention in this trial was as much as a publicly funded system is likely to commit. The teaching competence of those delivering the intervention in the trial was assessed at an average of “Advanced Beginner” on the MBI:TAC, the lowest end of the range of acceptable competence. In the previous smaller scale trial, the quality of the teaching was high, and the credibility of the intervention with the school children was high. ⁶⁹ In contrast, the credibility of the intervention was generally low in the MYRIAD trial, with higher levels of credibility being linked with higher levels of teaching competence. ⁷⁰ Trials that prioritize evaluation of effectiveness would select teachers who are at least “Proficient” and have a depth of experience. MYRIAD focused on effectiveness within the context of a pragmatic evaluation of a realistic implementation strategy. The results indicate that this curriculum cannot be scaled up by training teachers who are new to mindfulness and the curriculum. The MYRIAD trial points to the importance of explicitly building in process analysis of implementation factors into research, and ideally into delivery in routine practice contexts. There are many factors beyond the intervention and the teacher that influence the effectiveness of the MBP. ⁷¹ It is important to test whether the beneficial outcomes seen in research trials can be delivered in the pragmatic realities of mainstream settings.

The Greco et al. Klatt et al, ⁴⁷ and Serpa et al ⁴⁴ articles in this collection are examples of research examining the processes involved in scaling up MBP delivery. The two papers on the experience of implementation in the Swedish context illustrate research projects that foreground questions related to implementation.^48,49 The field needs more of these trials asking pragmatic implementation questions going forward.

The Flexibility-Fidelity Debate

There is a long debate in the literature on implementation on whether interventions should be implemented with maximum fidelity or whether adaptation should be encouraged to tailor the delivery to the context and population. ⁶³ Similarly, in the MBP field there is debate on whether to aim for breadth (ie, maximum accessibility across the population), or depth (ie, enabling more thorough and detailed engagement with learning). ⁷² This has become a particularly intense debate in the context of the rise in digital delivery. In practice, these are not “either-or” dichotomies. Rather the focus should be on finding the right balance of adaptation, and the optimal depth and delivery methodology for particular groups and individuals. The Loucks et al paper ³² in this collection unpacks how theoretical understanding of the core components of an intervention can be translated into the work of adapting programs for different contexts, cultures, and populations. Gaining empirical clarity on what elements of intervention fidelity are most important for participant outcomes would inform future research and practice by identifying critical intervention components to prioritise. We need additional research to establish predictive validity of MBP fidelity measures and identify which elements may be most important for the participant experience and outcomes. The limitation in existing data remains an important challenge in the field. However, as the Crane et al ³⁴ paper in this special issue unpacks, an expanded view of fidelity that is inclusive of the importance of contextual and cultural adaptation is completely consistent with the ethos of MBP pedagogy. Adaptations that improve engagement (ie, through alignment to values, cultural meanings, and practices) are highly likely to make an intervention more effective in a specific context. A shadow side of the completely appropriate emphasis on fidelity in MBP delivery could be a fear of compromising the quality of the mindfulness element of the intervention, which leads to a potential rigidity about “right and wrong” curriculum elements and processes in MBPs.

Next Steps for the Field

Research on Intervention Fidelity and Teaching Competence

MBPs are complex, multicomponent interventions. Without steps to assess intervention fidelity that can be reproduced in future work, it is difficult to convey to other researchers and practitioners exactly what was done in the program, and what level of teacher skill was involved. Without reproducible assessments of intervention quality and fidelity, it is impossible to know whether failure to replicate research findings is due to differences in program quality and fidelity, or other differences such as variations in study population. Being able to assess intervention fidelity in real world settings is also potentially critical to ensuring dissemination of research findings in an effective way.

Until recently, efforts to achieve intervention fidelity have focused on two steps: (1) developing a detailed manual of the intervention that allows replication; (2) Is to assess whether elements of the curriculum felt to be important are implemented as planned. While we feel these steps are potentially important for intervention fidelity, they may leave an essential element of intervention fidelity unaddressed: teacher skill in conveying mindfulness. The limitations of these approaches were underlined in the results of a randomized, controlled trial one of us (FMH) did of adding mindfulness components to diet and exercise alone for weight loss in people with body-mass index between 30 and 45. ⁷³ The mindfulness components were drawn from MBSR and Mindfulness-based Eating Awareness Training and the program was carefully manualized in a document that was over 200 pages. During the trial, one teacher followed the manual more carefully than the other two teachers, who were more flexible about following manual guidelines, and, for example, would pursue more dialogue with participants to explore challenges and experiences when the teacher felt this was productive, even when this went over suggested times for discussion. Although this was a post-hoc analysis that must be regarded with caution, the two teachers who followed the manual less closely had statistically significantly better weight loss in participants compared to the teacher who was most adherent to the manual, and these two teachers were rated as more helpful by participants. These results suggest that detailed manualization alone may not ensure that an MBP is delivered in an effective way, and that other steps are needed to ensure fidelity.

Where do the papers in this collection, and other available literature leave us in assessing intervention fidelity in MBPs using tools like the MBI:TAC? A few conclusions can be reached. An initial question in assessing the MBP teaching skill is whether there is a reproducible tool that has good inter-rater reliability for something as difficult to quantify as skilful mindfulness teaching. The collective results of research so far indicates that the MBI:TAC, used by a trained expert, generally has good inter-rater reliability. Several cautions, however, are notable from this collection. Based on the findings in this collection and prior data, we suggest the following practices for optimal implementation of an tool like the MBI:TAC in research studies:

• Assessors need to be adequately trained in the use of the tool, including showing that they are able to use the MBI:TAC in a way that correlates well with gold standard expert assessments.

• It is optimal to have more than one assessor assessing each MBP teacher for good inter-rater reliability.

• It is also optimal for assessors to review at least two course sessions when making an evaluation of an intervention like MBSR.

• Audio recordings may have adequate inter-rater reliability for some purposes, but including video is optimal and improves inter-rater reliability.

What else needs to be done in the field? The articles in this collection point to important limitations in our current knowledge of intervention fidelity for MBPs in research settings. While the MBI:TAC may have the strongest data supporting inter-rater reliability when evaluating teacher skill and has preliminary predictive validity data to support it, there are also important limitations that we have noted. Other tools reported in this paper require less time, but focus on adherence, and do not generate nuanced assessments of competence.

Areas that need further study include:

• We need further data on the predictive validity of tools such as the MBI:TAC, with a focus on how well they capture teaching skills that are important in providing particular benefits for participants. While the PrOMPT study suggests that teachers that are assessed as more skilful on the MBI:TAC may provide greater benefit to participants in reducing anxiety, other measures, such as depression, have not yet been shown to predict greater benefit in the PrOMPT study, or in the other prior study that has tested predictive validity of the MBI:TAC. ⁷⁴ This work requires greater numbers of participants, potentially in clinical populations (in which there are fewer floor effects for outcomes such as depression or anxiety), and is an important future research task.

• We need further data on effective ways to simplify the use of tools like the MBI:TAC for studies in which the use of multiple expert assessors is difficult. One option would be to make greater use of teacher self-assessments. The Serpa et al. article, ⁴⁴ however, suggests that self-assessment using the MBI:TAC results in assessments do not have good correspondence to those of an expert assessor. Another potential tool is participant assessments of their teacher. While this is potentially promising and work is being done to develop such a tool, further data assessing this approach is needed. The Floyd et al article suggests that one way of simplifying is to use audio recordings rather than video recordings to make assessments, but whilst this can work in some circumstances, there is some loss of inter-rater reliability. ⁴¹

• As noted earlier, instruments like the MBI:TAC focus on teacher skill, whereas other instruments, such as the CoFi-MBI and the fidelity assessment approach described by Greco et al, both described in this special collection, include a greater emphasis on adherence to delivering specific components of an MBP curriculum. We need further research on the relationship between these two types of approaches (how much are they correlated vs independent of each other in practice), their respective predictive validity, and the utility of combining instruments that focus on teacher skill, like the MBI:TAC, with checklists for specific intervention components to potentially achieve a broad assessment of intervention fidelity.

• We also need further research on how to best modify fidelity tools for use in adapted MBPs that differ from the focus of their original development and testing.

Conclusions

The international disruption of the pandemic resulted in this special collection being open for submissions over a longer period, which has revealed some shifting emphases in MBP research and practice. MBPs already had a solid basis of research and practice addressing individual health and wellbeing. The long-standing global challenges of social inequity and polarization, and the climate and biodiversity crisis have become more pressing issues through the pandemic years, and this is shaping and expanding the MBP field with more researchers and practitioners tuning their work towards responding to these themes. How the field navigates the tensions inherent in honouring the evidence base and fidelity of established MBP forms, whilst enabling quality innovation, and inclusion of contributions and expertise from those who have been historically marginalized are vital questions shaping the direction of this work. We hope that this collection will increase dialogue about the strengths and limitations of the MBI:TAC and other available tools for assessing teaching skill, and how to strengthen their use in the future; and that it will stimulate creative engagement with the need for complementary fidelity tools and methods and will catalyse interest in the research community in examining the complex, multifaceted, and infinitely fascinating questions about MBP curriculums and teaching processes. Although important advances have been made in developing reliable methods for assessing MBP intervention fidelity that address teacher skill, there are clearly key remaining research questions for the field, including better establishing the relationship between fidelity measures and participant outcomes. Questions about MBP intervention fidelity are at the meeting place between research and practice. It is an important developmental step for the field to bring the focus of research inquiry closer to the processes that are bringing about the beneficial outcomes we are seeing. We hope this engagement will stimulate the MBP teaching and research communities internationally to collaborate on these questions to optimise the benefits of this promising field going forward. Developing validated fidelity measures that are based on empirically founded understandings of the active ingredients of mindfulness-based program content and process, and elements of teacher skill and what is needed to optimise participant benefit, are vital foundations for enabling this promising field to evolve and adapt to meet changing contexts and new populations.