Summary and discussion of the 10 papers in this Special Issue

Paper 1: Supporting our supervisors. In their introductory review paper, Milne & Reiser (Reference Milne and Reiser2016) propose a novel ‘supporting our supervisors’ (SOS) framework, consisting of seven actions that they regard as a procedural blueprint for attending effectively to CBT supervision at the organizational level. The seven actions are:

• • conceptualization (formulating the supervision system, to understand the local issues);

• • operationalizing (measuring key support factors, such as staff burnout or peer support);

• • data collection, to quantify specific challenges (e.g. irregular provision of supervision within a clinical service);

• • interpretation of local data and of published research (e.g. does supervision address supervisors’ support needs adequately?);

• • dissemination of promising approaches (e.g. guidelines on supportive supervision);

• • utilization of support and development approaches (e.g. supervisor training, drawing on such guidelines); and

• • evaluation, to monitor and enhance support arrangements (e.g. is the supervisor training effective?).

This SOS framework is intended as a useful heuristic to guide and organize efforts towards a more coherent and systemic approach to supporting supervisors. We reiterate the optimism embedded within Milne & Reiser's (Reference Milne and Reiser2016) SOS framework. Gone are the days when training in supervision was neglected, taken for granted as an instinctively acquired competency, unmeasured and untested, and rarely represented effectively in the literature (Watkins & Wang, Reference Watkins, Wang, Watkins and Milne2014). There is also a growing body of research on evidence-based CBT supervision, and a new emphasis on state-of-the-art training, including professional standards, manuals, and measurement tools with which to further develop this critical area of our profession. We foresee continued progress in the future, including a sustained focus on research, training, and practice in CBT supervision at professional conferences. This will encourage dissemination, promote research collaboration, and provide valuable opportunities for cross-cultural communication and adaptations. With regard to training programmes for therapists, we anticipate regular supervisors’ meetings, together with providing these supervisors with the material support that they need (e.g. the systematic collection of trainees’ evaluations of supervisors, so as to provide vital feedback). Within organizations like the British National Health Service (NHS), we anticipate that supervision sessions will more frequently be recorded for supervision-of-supervsion, future training purposes, and routine data collection. Technological advances will permit more widespread access to training via video-linking, including both individual and group supervision, along with improved methods of tracking and flagging clients’ progress for better oversight and reduced risks. Accreditation bodies evaluating therapists’ professional training programmes will increasingly assess the preparation that trainees receive for being a supervisee and a supervisor. For established clinicians, maintaining one's professional registration or licence to practice will necessitate confirmation of taking part in periodic high-quality training in supervision, such as those described in this Special Issue. Organizations and service managers will also have to play a greater part than before in this much more supportive context for supervisors.

Paper 2: A survey of CBT supervision in the UK. Further specific evidence on how to support supervisors was next reported by Reiser & Milne (Reference Reiser and Milne2016), based on a survey of 110 CBT supervision leaders in Britain. The survey indicates that self-reported supervisory practices are consistent with current competency recommendations in the field (e.g. directly observing supervisees' work with clients and using a range of experiential training methods in supervision sessions). However, supervisors require improved supervisor training materials in order to maintain momentum in developing CBT supervision.

Thankfully, several evidence-based and highly practical CBT supervision books have been published within the past year (e.g. Corrie & Lane, Reference Corrie and Lane2015; Sudak et al. Reference Sudak, Codd, Ludgate, Sokol, Fox, Reiser and Milne2016; Newman & Kaplan, Reference Newman and Kaplan2016) indicating that progress is being made with respect to supervisor training. Other signs that support for supervisors is growing include supervision training for trainees, which may involve mandatory coursework, including learning to use established observation scales to rate recordings of supervision (Newman, Reference Newman2013). For established supervisors there are now ‘self-help’ manuals (e.g. Corrie & Lane, Reference Corrie and Lane2015), and comprehensive programmes of supervisor development (e.g. the Advanced Cognitive Therapy Studies programme at the Oxford Cognitive Therapy Centre; https://www.octc.co.uk). We acknowledge that such developments require an increased supply of qualified trainers (Ferguson et al. Reference Ferguson, Harper, Platz, Sloan and Smith2016), an example of the greater managerial and institutional support that is necessary to enable the SOS model.

Trainers and supervisors can benefit from having scheduled opportunities for peer consultation. As Corrie & Lane (Reference Corrie and Lane2016) note, an excellent example of this model is the New Zealand Coaching and Mentoring Centre (2012). Consultation may also involve supervision-of-supervision, also known as ‘meta-supervision’, in which an experienced clinical supervisor observes, evaluates, and gives feedback plus guidance to a colleague on his or her supervisory work with a trainee (Newman, Reference Newman2013). As noted by Buus et al. (Reference Buus, Lynch and Gonge2016), meta-supervision can be done in a group format, helping clinical staff to experience a sense of institutional support for their work (see paper 4, below).

Paper 3: Clinical supervision in a quality assurance/quality improvement system. Schoenwald (Reference Schoenwald2016) provides a comprehensive and powerful illustration of a system for supporting and maintaining quality and effectiveness in supervision, within the context of the large-scale implementation of Multisystemic Therapy (MST), an evidence-based treatment for at-risk youth. Schoenwald (Reference Schoenwald2016) describes how supervisors were supported through an impressive infrastructure, including the use of a manualized handbook of MST supervision, a formal standardized process for reviewing cases, and routine collection of therapist and supervisor adherence data on a validated measure. Supervisors were provided with periodic support from an expert MST consultant, including individualized feedback based on several samples of the supervisors’ work, and quarterly booster training sessions. There was also a consultant adherence measure, indicating that this was an exceptionally comprehensive infrastructure (covering all aspects of the SOS model), with compelling evidence of an effective quality improvement process. Not surprisingly, MST has been in development for many years, and the effort has required significant amounts of funding, energy, commitment, and collaboration.

Alongside Schoenwald, several articles in this Special Issue have noted the importance and benefits of giving feedback to supervisors (e.g. Corrie & Lane, Reference Corrie and Lane2016), and this is where measures of supervisors’ competence can play an important role in quality control. If delivered in an appropriately challenging and constructive fashion, even critical feedback can lead to professional growth (Rieck et al. Reference Rieck, Callahan and Watkins2015).

Paper 4: Developing and implementing ‘meta-supervision’. Before supervisors can benefit from such feedback procedures they need to be actively engaged in supervision. Gonge & Buus (Reference Gonge and Buus2010, Reference Gonge and Buus2011) have demonstrated low participation rates in supervision, illuminating some of the possible explanations. These include a systemic failure to prioritize supervision, anxieties about exposing work-related concerns in supervision groups, and difficulty transferring material from supervision to everyday practice. To address these challenges, Buus et al. (Reference Buus, Lynch and Gonge2016) examine a group-based, ‘meta-supervision’ intervention, led by an external ‘meta-supervisor’. Meta-supervision is designed to promote ownership, morale, self-reflection, and individual problem-solving on the personal and organizational barriers to engaging in supervision. Data from their pilot study suggest that participation in supervision increased as a result of their initiative, although there was reported to be little improvement in the formative and restorative benefits of supervision. This paper illustrates a thoughtful consideration of internal cultural and organizational barriers to supervision, including such simple but discouraging factors as failing to hold regular team meetings, thus giving the message that staff support is expendable.

The target group for the Buus et al. (Reference Buus, Lynch and Gonge2016) study was comprised of nurses working in residential practice. Thus, the meta-supervision described in their paper contrasts with meta-supervision as described in outpatient CBT in the clinical psychology literature (e.g. Armstrong & Freeston, Reference Armstrong, Freeston and Tarrier2006; Newman, Reference Newman2013). Nonetheless, there are important similarities, including the benefits of giving clinicians extra feedback on their work, a supportive professional relationship, encouraging self-reflection, and more active participation in their own training.

Paper 5: Organizing professional development to enhance practice. Like Buus et al. (Reference Buus, Lynch and Gonge2016), Corrie & Lane (Reference Corrie and Lane2016) recognize that supervisor development should not be viewed solely as a personal learning process that is left to individuals, but in order to succeed needs to contextualize supervision within the organizational environment. At the same time, Corrie & Lane's four-step PURE model (Preparing, Undertaking, Refining and Enhancing supervision) provides well-structured guidance to anyone who wishes to conduct individual or group CBT supervision, or to train supervisors. The PURE model is a visual heuristic in the form of a flower containing 12 ‘petals’, each of which is concerned with a specific domain of activity the CBT supervisor needs to master, and three of which are subsumed under each of the four steps. For example, the petal ‘negotiating a contract’ is in the preparing step, ‘assessing trainee competence’ is part of the undertaking, ‘resolving ruptures’ is a manifestation of refining, and ‘increasing supervisor expertise’ comes under enhancing. Their supervision manual (Corrie & Lane, Reference Corrie and Lane2015) contains open-ended exercises that assist supervisors at each step in the PURE model, allowing supervisors and their trainees to think out loud about how to solve problems and maximize learning. This approach is very much in the spirit of ‘guided discovery’, one of the key elements of competent CBT, and one of the scoring items in the CTS and the CTS-R.

The authors recognize that today's supervisors face daunting tasks, such as needing to be conversant with a wide range of problem-specific treatment protocols, and having to operate within economically mandated (rather than research-informed) time limits. Supporting our supervisors means that we must recognize and address such work challenges, or further problems may arise within the clinical service, such as those noted above in the Buus et al. (Reference Buus, Lynch and Gonge2016) study. Other likely consequences are avoidance of supervision, and fatigue that can lead to ‘drifting’ from established procedures (Waller, Reference Waller2009).

Paper 6: Video feedback in CBT supervision. Complementing the above systemic papers, several contributors to this Special Issue provided specific technical, educational or procedurally oriented accounts of how supervision might be enhanced, primarily through supervisor training. For example, Gonsalvez et al. (Reference Gonsalvez, Brockman and Hill2016) offer a detailed illustration of the use of video material within such training, based on learning exercises that encouraged supervisors to provide two different types of formative feedback to supervisees in order to enhance reflection, and so build on supervisee competencies. The procedures for the learning exercises are carefully spelled out, with case examples and vignettes.

Gonsalvez et al.’s explicit instructions help to provide a pragmatic set of techniques for supervisors wishing to standardize and enhance their use of observational procedures. Their contribution is consistent with the empirically supported view that observational techniques are particularly helpful in addressing skills, attitude-value, and relationship competencies, offering an important alternative to the overuse of subjective, exclusively verbal case reports in supervision. Gonsalvez et al. (Reference Gonsalvez, Brockman and Hill2016) also indicate that the use of video can provide supervisors with any number of replicable opportunities to create engaging teaching points that trainees will find memorable. For example, Gonsalvez et al. (Reference Gonsalvez, Brockman and Hill2016) state that video review allows for focusing on process issues in therapy. To illustrate, a supervisor could show a recording of his/her own work with a client in group supervision, pausing the video at key moments, and asking the group, ‘Based on your conceptualization of this case, what do you think is going to happen next, and what would you do if you were the therapist in this situation?’ The supervisor then solicits thoughtful answers from each trainee in the group and then says, ‘Let's see what actually happened’, whereupon he/she resumes the video, which should stimulate attention and learning. Such a use of video, repeated many times, will undoubtedly serve as excellent training in self-reflection, a key component in CBT supervision (Bennett-Levy et al. Reference Bennett-Levy, Thwaites, Haarhoff and Perry2015).

Paper 7: How can video recordings best contribute to clinical supervisor training? The value of video material in supervisor training was also recognized by Milne (Reference Milne2016), who reviews the extant video materials, together with the way that such video material has been utilized within workshops. Like Gonsalvez et al. (Reference Gonsalvez, Brockman and Hill2016), he encourages a more systematic and evidence-based approach, noting that video material has been used in a restricted way educationally, and is also generally lacking in procedural detail (leaving video use entirely to the discretion of workshop leaders), a training rationale, or research support. By utilizing an approach constructed around competency frameworks for supervision, Milne suggests a constructive and methodologically sound advance in the use of video learning materials.

Sudak et al. (Reference Sudak, Codd, Ludgate, Sokol, Fox, Reiser and Milne2016) have also linked their text closely to a video-library that includes the modelling of such supervisory skills as working effectively with supervisees who have negative beliefs about CBT, or who have skill deficits. In addition, the video that serves as a companion to the Newman & Kaplan (Reference Newman and Kaplan2016) text on CBT supervision models the interweaving of cross-cultural challenges and ethical problem-solving into an otherwise standard supervision session. In the future, we envisage that the use of such videos will become more widespread, more routine, and more specifically targeted to supervisor competencies and training needs. Naturally, the effectiveness of such videos in improving supervision will need to be determined, and thankfully there are now a number of psychometrically sound instruments. These include the observational tool SAGE (Supervision: Adherence, and Guidance Evaluation; Milne & Reiser, Reference Milne, Reiser, Watkins and Milne2014), the Supervision Evaluation Scale (SES; Corrie & Worrell, Reference Corrie and Worrell2012), and the Supervision Competence Scale (SCS; see Rakovshik, Reference Rakovshik2015).

Paper 8: Developing specialist CBT supervision training. Ferguson et al. (Reference Ferguson, Harper, Platz, Sloan and Smith2016) describe the development, implementation and evaluation strategy for a blended learning programme for NHS supervisors in Scotland, based on the Roth & Pilling (Reference Roth and Pilling2008) framework of generic and speciality supervisory competencies. The term ‘blended’ refers to the fact that this programme combined the flexibility of an e-learning format, designed to combine interactive learning exercises with reflection, with an on-site workshop that focused on competency frameworks, supervision models, the development of a personalized learning plan, and role-play practice. Finally, a post-workshop e-module evaluated the participants’ learning, through a video clip and an associated multiple choice quiz. The evaluation strategy included learner satisfaction ratings, self-assessment of competence, the e-learning module quiz, and a 3-month follow-up (by email) assessing the transfer of training. The authors have found that the blended learning approach results in somewhat better outcomes than training in classrooms alone.

When considering the effort that blended learning entails, it is worth pausing to consider in general how well we support our supervisors through positive reinforcement. Recognition, promotion, financial gain and other reinforcers should form part of the contribution that managers, administrators and organizations make. However, in the UK and the USA, clinical supervision is not always compensated, and does not always count as an asset when the supervisors are considered for clinical or academic promotion, especially in large organizations and medical institutions. Supervision is an ethical and academic necessity when training therapists who are not yet qualified (licensed to practice independently), and in the UK it is essential throughout therapists’ careers. But in many settings supervision may be considered a luxury, and providing supervision devalued because it does not bring in revenue to the organization (e.g. classed as ‘unused hours’ on the accounting sheets). By contrast, the SOS model requires that we provide supervisors with the kind of reinforcement that we see in projects such as described by Schoenwald (Reference Schoenwald2016), in which the clinical team has a personal stake in positive outcomes for trainees and clients alike. Moreover, the quality assurance system for Multisystemic Therapy perhaps comes closest to demonstrating how to achieve support at the highest administrative levels, and to gain ‘buy-in’ from policy-makers, so that a promising programme of supervision and therapy can become a fixture in a community mental health system.

Paper 9: Training supervisors to use direct assessments of clinical competence in CBT. Loades & Armstrong (Reference Loades and Armstrong2016) examine another training-related challenge: establishing inter-rater reliability amongst supervisors who assess therapist competence (i.e. applying the CTS and its revised version, the CTS-R). In their review of 20 inter-rater reliability studies that reported data on these two scales, the authors find significant variability in reliability both between studies and across scale items. They express concern about the lack of procedural data on the reliability training of the raters (supervisors), with a majority of studies omitting information about the type, form, frequency and duration of the training. In their own exploratory study they were able to rectify these omissions by detailing their own training procedure. However, there were disappointing results: raters reported improved confidence in their ratings on the CTS-R, but there was no improvement in the reliability or accuracy of their ratings by the end of the one-day workshop. This study illustrates the challenges in improving inter-rater reliability on these key competence measures, especially when training time is limited. Given that these two scales help supervisors fulfil their gate-keeping function, this is a major gap in our understanding and implementation of effective training.

We recognize that more work will be needed to achieve acceptable levels of inter-rater reliability, as there is a great deal of subjectivity involved in evaluating supervisors’ performance. In the meantime, we wish to note how valuable qualitative information can be, in conjunction with ratings, especially the supervisors’ suggestions on how the therapist/supervisee diverged from standard CBT practice, or how well the therapist managed unexpected difficulties. It is a notable strength that the instruments we have noted all include sections for such feedback. With this in mind, Ferguson et al. (Reference Ferguson, Harper, Platz, Sloan and Smith2016) suggest that the CTS-R be used primarily to give trainees formative feedback rather than summative evaluations, so as to reduce the risk of penalizing trainees with low scores which may be attributable to unreliable ratings. In the same way, several UK programmes use SAGE purely formatively within their supervisor training, including qualitative comments (e.g. suggestions on improving the observed supervision). We do not know of any studies, regarding the instruments that we have mentioned, that have included an evaluation of such qualitative data. It may be that such an analysis would aid our understanding of the recurring challenge of low reliability between raters. As a case in point, challenging clients are prone to have therapy-interfering behaviours that compel a responsive therapist to go off strict protocol in order to set limits, repair the therapeutic alliance, maintain proper boundaries, and keep at-risk patients safe (see Zickgraf et al. Reference Zickgraf, Chambless, McCarthy, Gallop, Sharpless, Milrod and Barber2016). In order to study the links between supervisor behaviour, therapist behaviour, and client outcome, our measures (and training methods) need to take into account such real-life vicissitudes of clinical practice (Stirman et al. Reference Stirman, Calloway, Toder, Miller, DeVito, Meisel and Crits-Christoph2013).

Paper 10: Technology for evidence-based cognitive behavioural supervision. Placing training issues in the context of the internet, Rousmaniere (Reference Rousmaniere2016) identifies two important technological advances that provide support for supervisors: software for routine outcome measurement (ROM), and the development of a remote live supervision application. While the use of standardized instruments for monitoring progress has been a mainstay of supervision and clinical practice in CBT, these new technology applications simplify routine outcome monitoring and offer rapid availability of outcome data in graphic form. Of particular interest is the development of outcome tracking software, intended to provide an overview of clinical decision-making support at an organizational level. This type of data management system has tremendous potential value in assisting supervisors and managers in profiling cases, including assessing risk factors rapidly, and triaging supervision in order to be as effective as possible. With higher demands on therapists and supervisors, such as the IAPT low intensity services, we believe that this type of decision support is going to be increasingly critical to effective supervision. Rousmaniere (Reference Rousmaniere2016) pairs this overview of software with a discussion of remote live supervision technology, which offers the benefits of direct real-time observation to staff at geographically distant and/or hard-to-reach locations. This technology provides immediate feedback to supervisees, either through text-based interventions (appearing on a computer screen visible only to the therapist), or by using a ‘bug-in-the-ear’ method of live supervision.

As Rousmaniere (Reference Rousmaniere2016) notes, live video-linked supervision will surely become more prevalent, creating widespread training opportunities. The American Psychological Association has published a statement called the ‘Guidelines for the Practice of Telepsychology’ (APA, 2013), in which clinical supervisors are encouraged to continue in-person supervision as the preferred choice (when possible), to be reasonably proficient in the technologies they are using, and to consult with knowledgeable others to reduce risks regarding the security of client data. The use of video-linked supervision at a distance also raises the question of how the lack of the supervisor's physical presence will impact the quality of the trainee's learning experience. Fortunately, there are data suggesting that ‘cyber-supervision’ is as effective as individual face-to-face supervision, and that both supervisors and supervisees have perceived no significant differences in the quality of the supervisory relationship (Inman et al. Reference Inman, Hutman, Pendse, Devdas, Luu, Ellis, Watkins and Milne2014).