An Introduction to Cognitive Behaviour Therapy, 2 edition


Using Supervision in CBT


As anyone who tries it soon discovers, the practice of good therapy is not something you can learn quickly or easily. You cannot just read a book, or attend a workshop, and then go off and do good CBT: effective clinical training demands a much longer process of learning, in which you bring together what you have learned about theory and therapeutic strategies with the complex reality of the clients you see. Clinical supervision is one of the main ways in which this continued learning can occur. It can take different forms (see below), but the basic idea is that by having someone else to discuss or directly observe your therapy, you can examine how well it is going, identify problems, find solutions and develop your skills. The need for supervision applies to CBT practitioners at any level, but most crucially for novices.

These views of the positive value of supervision would probably gain support from most CBT practitioners. However, embarrassingly for an approach that values empiricism as CBT does, there is not yet much evidence about whether supervision actually makes a difference, either to the supervisee’s skills or to the outcomes for clients. There is one important recent study (Mannix et al., 2006) that found that clinicians receiving supervision after training in CBT did better in maintaining and improving their CBT skills than people who did not get supervision, but much more evidence is needed. What follows is, therefore, based mostly on the clinical experience and beliefs of ourselves and others, rather than on a strong evidence base. Much of what we have to say can and should be challenged as new evidence emerges.

Goals of supervision

Although we are unaware of any single generally agreed definition of clinical supervision in CBT, there is agreement that clinical supervision can help achieve any or all of the following goals:


  • developing therapist skills: honing and improving existing skills, and learning new skills;
  • protecting clients: providing a form of quality control for therapy, both at the practical level of ensuring appropriate strategies are being used and at the emotional level of enabling an external, more objective view of therapeutic relationships;
  • providing support for therapists in dealing with the difficulties therapy may cause for them;
  • monitoring and evaluating therapist skills and practice.

The balance between these different components will vary according to factors such as the characteristics and experience of therapist and supervisor, the context of the supervision and so on.

In thinking about the last of the goals, i.e. evaluation, it is worth considering the distinction between summative and formative evaluation.


  • Summative evaluationrefers to evaluation whose primary aim is to arrive at a summary judgement: is the subject of evaluation ‘good enough’ in some sense (e.g. is the trainee good enough to pass the course)?
  • Formative evaluationrefers to evaluation whose primary aim is to help the subject to improve, i.e. where the main point is not ‘Is X good enough?’ but rather ‘How can we make X better?’ Almost all clinical supervision contains elements of formative evaluation, but summative evaluation is usually only important in the context of a training course, therapist accreditation, or similar processes.

A useful tool to evaluate cognitive-therapy skills for both summative and formative purposes is the cognitive therapy scale (CTS: Young & Beck, 1980; Dobson, Shaw & Vallis, 1985; or a revised version, the CTS-R – see Blackburn et al., 2001).

Modes of supervision

We can distinguish two important dimensions of supervision: first, whether the supervision is for an individual therapist or a group; second, whether the supervision is from one person considered to be more expert to another considered less expert or is between people of roughly equal expertise. Combining these categories gives us four modes of supervision, which we have given arbitrary names, illustrated in Table 19.1.

It is sometimes thought – wrongly in our view – that all supervision must have an identified leader. We feel that peer supervision can be very useful, even if none of the participants is highly expert. We would draw an analogy with CBT therapy: just as the therapist may be able to help the client in an area about which she knows little, through a process of guided discovery, so peers may be able to use the same processes to open up new ways of thinking for the therapist in supervision. Although there is a danger of ‘the blind leading the blind’ if none of the participants has any CBT experience, such supervision can be better than nothing at all in situations where access to expert supervision is limited.

Table 19.1    Modes of supervision

There are pros and cons to each of these modes.

1 Apprenticeship

This is probably what most people think of as typical supervision: a skilled and experienced practitioner meeting a relative novice, one to one, to develop the novice therapist’s skills. It is undoubtedly a good model, with excellent scope for detailed examination and rehearsal of therapy skills, finely tuned to the supervisee’s needs. The main disadvantages are that it is relatively extravagant in its use of the leader’s time (and thus expensive); and because there is only one supervisor, the range of views and expertise available to the supervisee is limited.

2 Led group

The main selling point of the led group is that it has the advantages of the apprenticeship model in terms of offering expertise, whilst being more economical and therefore more practical in many settings. Another advantage is that supervisees can learn from hearing about other practitioner’s cases as well as their own. Possible disadvantages include the fact that each person in the group gets less individual time and that it can sometimes become more like a seminar – although this too can sometimes be helpful.

3 Consultation

We use this term to refer to a similar set-up to apprenticeship – i.e. two individuals meeting for supervision – but in this case they are roughly equally skilled, so neither one is an ‘apprentice’. For very experienced practitioners, this may well be the only available mode of supervision, since there may be no one with more expertise available. Consultation can either be one-way – one of the pair consults the other for supervision – or can be mutual, where each supervises the other.

4 Peer group

The advantages of the peer group include its being relatively cheap and easy to set up; that it allows vicarious learning; and that by being more egalitarian it may encourage less experienced participants to be creative and to share their ideas. Disadvantages include the risk that ‘the blind can lead the blind’, with no one really knowing what they are talking about; that, as with any form of group supervision, each supervisee gets less time; and that there is no leader to take responsibility for group dynamics.

Alternative channels for supervision

In addition to the above modes, it is worth considering that some of them can be practised using other means of communication, such as telephone and e-mail. These alternative channels probably work best with individual supervision: managing group interactions without any face-to-face contact is not easy! Such methods may lose some of the subtleties of face-to-face communication, especially when it comes to emotional issues, and there are possible technical difficulties in playing audio- or video-tapes. Nevertheless, they can offer a useful alternative if no supervisor is available to meet your needs.

The authors have several supervisees who get clinical supervision from us by phone (usually because they have no appropriate supervisor within a reasonable travelling time). In fact in some cases we have never actually met the supervisee face-to-face. Such supervisory relationships can work well despite the restrictions of this format. As with any mode that excludes some non-verbal aspects of communication, the potential for misunderstanding is increased, so it may need particular attention from both parties to ensure clear summarising and feedback in order to avoid such problems. A more recent innovation is doing supervision over the Internet, using webcams so that the parties can both see and hear each other. Some such systems even allow tapes to be played. This approach can work well, but it does need reasonably powerful technology and a fast internet connection at both ends.

Supervision or therapy?

It is recognised in all forms of therapy that there may be times when the therapist’s own problems or beliefs may impinge on therapy. Is such material suitable for exploration in supervision, or should it be considered as personal therapy that should take place in another setting? Is there a limit to the personal material dealt with in supervision sessions? If so, where will you go with material that it is agreed should not be part of supervision? There are no right answers to these questions, but most CBT practitioners would say that such material should only be part of supervision to the extent that it has a direct impact on your work with your clients (e.g. Padesky, 1996b). If one of your own beliefs (for example, ‘I must never do anything that clients will find distressing’ or ‘I am solely responsible for my clients’ progress’) is blocking you from implementing a therapeutic strategy that appears relevant in other respects, then it may be appropriate to look at that in supervision. If it appears that the belief is part of a wider problem, it probably cannot be dealt with in the limited time available in supervision. If it is not something that directly affects your work with patients, then it should probably be taken elsewhere.

Use of tapes

The use of audio or video recordings of client sessions (just ‘tapes’ from here on, although of course many such recordings are now digital) has always been a distinctive feature of CBT supervision. Instead of the supervisor having to rely on the supervisee’s account of therapy, the supervisee records sessions and plays back (parts of) the tapes during supervision so that the supervisor can observe what happens more directly. Although almost everyone initially feels anxious about presenting their therapy openly to others in this way, it gets easier and is extremely useful once the initial inhibitions have been overcome. We therefore strongly advise using tapes. Advantages include:


  • Self-reflection. Although not always comfortable, it is good practice to listen to your therapy tapes and critically appraise your performance. In addition, you can better prepare your supervision question and identify the most relevant parts of the tape to share with your supervisor.
  • Avoiding omissions and distortions(positive or negative) in the therapist’s view and account of clients. It could be that what is important in the sessions is something you have not noticed or are reluctant to report – and tapes give you and the supervisor a chance to spot that.
  • It therefore allows a depth and precision of supervisionthat is almost impossible if supervision is based only on the therapist’s account. Instead of an inevitably partial account from the therapist, the supervisor can hear or see more of the full complexity of the interaction. Think how long it would take you to give a detailed verbal description of a therapeutic interaction, compared to what can be gleaned from listening to a couple of minutes of tape! It is difficult for most people to convey a full sense of what a client is like through verbal description. Via tapes, the supervisor can get a much better impression of your client, as well as what is happening in the therapy, and may therefore be better able to offer useful suggestions for how to manage the client’s behaviour.
  • If you are intending to work towards accreditation as a therapist by the British Association for Behavioural & Cognitive Psychotherapies (BABCP), then bear in mind that a certain amount of what they call ‘live supervision’ (i.e. your supervisor hearing or seeing actual client sessions, directly or through recordings) is now a requirement. See www.babcp.comfor more information about this.

If you are using tapes then you need to consider the practicalities of how to do so effectively and ethically:


  • Clients must give full informed consent before any sessions are recorded. You therefore need to establish processes to obtain and record the client’s consent. Your hospital or other organisation may have a policy that you need to follow. If not, you should think about developing a policy of your own which includes informing clients what will be done with the tapes, who will hear them and what arrangements will be made for storing them, and for destroying or securely wiping the tapes after supervision. Many NHS agencies nowadays demand that digital recordings are kept in an encryptedformat, so that they cannot be listened to by anyone else if they are lost or stolen.
  • If you have decided to use tapes, then it is usually much easier if you get into the habit of recording allyour client sessions rather than thinking about it only when something has gone wrong or a specific need has arisen. It is usually much easier to get clients’ consent to a routine procedure at the start of therapy rather than negotiating recording with the client when some particular difficulty has come up.

Choosing a supervisor

In some situations, particularly when training, you may have no choice about who your supervisor is because he or she is allocated to you, but if you have a choice then you may want to consider the following:


  • Is the supervisor someone you think you can trust and be comfortable with? You need to be able to form a good working relationship that is as important – and sometimes as difficult – as a therapeutic relationship.
  • Does the supervisor have the skills that you need to learn from? For example, you might want someone who is expert in treating certain problems, or someone who can work more with the therapeutic process. It may be difficult to judge initially how well someone will meet your needs, hence the importance of agreeing an initial trial period (see below).
  • Is the supervisor motivated to supervise you and willing to commit to the agreed period of supervision?
  • Is the supervisor receiving supervision? It is increasingly common for supervisors to seek supervision of their practice.
  • BABCP now offers a pathway for accredited therapists to be accredited as supervisors as well. The number of people with this additional qualification is relatively small at the time of writing, so having an accredited supervisor is not yet a requirement, but you may want at least to enquire whether a potential supervisor is accredited as such. Roth & Pilling, whose competence framework for CBT we mentioned in Chapter 1, have developed a similar set of competences for supervision, which you or your supervisor may find of interest (Roth & Pilling, 2009).

Negotiating supervision arrangements

Whether you are choosing a supervisor or have one allocated to you, it is always important to have a preliminary meeting in order to clarify hopes and expectations. You may want to consider:


  • Practical arrangements: when supervision will take place, where, how long will sessions be, how often, etc.?
  • Confidentiality issues.
  • Is the supervision to be ‘general’ or are there specific objectives (e.g. to get better at formulating, or dealing with clients with OCD, or managing clients who are self-harming)? If you have specific objectives, does the supervisor feel she has the necessary skills to help?
  • Do you or the supervisor have priorities regarding the possible aims of supervision outlined above? For example, are you on a training course where summative evaluation is an important part of supervision?
  • Although it is often impossible to predict in advance what will happen during a course of supervision, it is worth considering your views and the supervisor’s about the boundaries between supervision and personal therapy.
  • It is usually wise to have a trial period and then review so that if it is not working well the arrangements can be changed.

See the Appendix to this chapter for an example template for a supervision agreement, provided by the British Association for Behavioural & Cognitive Psychotherapy (BABCP, 2005).

For any kind of group supervision, similar questions will arise, and there will also be others unique to the group setting. For example, how will the available time be divided between the group members? Options include equal time for all group members in every session; or one person presenting at each session (with possibly long intervals between supervision sessions for any one individual); or a combination of these, in which one person takes a larger chunk of time in a session but others also take smaller amounts of time. There may be more need for explicit feedback from the supervisee(s) to the group and/or the leader about their experience of supervision. If, as sometimes happens, most or all of the group members disagree with a supervisee’s approach, the supervisee can be left feeling very exposed and isolated. Careful attention to group process may be needed to avoid excessively painful and off-putting experiences for supervisees.

Preparing for a supervision session

You will get the most out of supervision if you prepare carefully for it. This does not need to take hours, but it needs more than taking two minutes to think before supervision starts, grabbing the client’s notes and running off to meet the supervisor! In particular, we would recommend that you identify a reasonably clear supervision question for each client you will discuss. This does not necessarily mean a simple question but, rather, what it is that you want to focus on. The potential range of questions is limitless, but might include:


  • How can I develop a formulation for this client?
  • What might I do to manage this client’s tendency to get dependent on me?
  • What behavioural experiments could I usefully do with this client?
  • Can we look at what makes me get angry with this client?
  • Where can I learn more about treating OCD?

Preparation is also important in using tapes for supervision. It is almost impossible for supervisors to listen to complete tapes of all your clients, as that would take many hours. It is therefore best to make this a two-stage process. First, listen to your tapes yourself. This alone will provide you with useful insights about how you might improve. As you listen, make a note of any points where problems seem to emerge: those are the issues you might want to take to supervision. Select segments of a few minutes that illustrate the points you want to discuss, then, before the supervision session, have your tape ready to run from those points. With this kind of preparation, tapes can be used effectively and economically.

In preparing for supervision, Padesky’s (1996b) questions for supervisors can be helpful for supervisees as well. These questions include the following sequence of factors that might contribute to difficulties in therapy:


  1. Is there a CBT formulation and consequent treatment plan for this client?
    If not, developing one may be a supervision question.
  2. Is the formulation and consequent treatment plan being followed?
    If not, the supervision question might be thinking about what is stopping you from following the formulation and treatment plan – either something in your own beliefs or some characteristic or behaviour of the client.
  3. Do you have the knowledge and skills to implement the required treatment?
    If not, you might want to use supervision to gain knowledge, practise skills or get advice on where you can go to achieve those goals.
  4. Is the client’s response to therapy as expected?
    If not, you might want to use supervision to consider what client beliefs, life circumstances, developmental history, etc., might be blocking progress.
  5. If all the above have satisfactory answers, what else might be interfering?
    Supervision might need to consider therapist factors, problems in the therapeutic relationship, whether the formulation needs modifying, whether a different treatment approach is needed and so on. See Westbrook, Mueller, Kennerley and McManus (2010) for further thoughts about common difficulties in therapy.

Michael found it useful to go through this checklist in relation to a client with whom he was struggling. By doing so he realised that the problem seemed to be around Step 4: he had what seemed like a reasonable formulation and treatment plan, and he was following them, but it just did not seem to be working. The client’s dominant emotions were depression and anger, and she was preoccupied with justifying her anger, rather than working towards any of her other overt treatment goals. Having identified this point, Michael recognised that something seemed to be missing in the formulation, because this preoccupation with ‘justified anger’ was not currently adequately explained. Exploring further with his client, Michael realised that she had beliefs that meant that the only way she could assert her needs was to get angry; and that recognising that her anger might sometimes be excessive therefore seemed to her like subjugating herself to the other person. With this re-formulation, the patient was able to start working on other ways of asserting herself.

During a supervision session

The form of a supervision session can usefully follow the model of a CBT session, as outlined in Chapter 11. Thus, supervision might go through the same sequence:


  • Agenda-setting: what are today’s main topics and how will we divide the available time between them? Both parties should be able to raise topics for the agenda, and the supervisee should take increasing responsibility for the agenda as she becomes more experienced. Reviewing homework should always be on the agenda and there should be the opportunity to review the clinical consequences of the previous session’s supervision if this is not covered in reviewing homework.
  • Main topics: the bulk of the session will be taken up with going through the agreed main agenda items.
  • Homework assignment: supervisees will often leave supervision with some agreed list of tasks to be carried out, ranging from reading a particular article to trying out a particular strategy with a client.
  • Review: what feedback does the supervisee have about the session, what has she learned, what was difficult and so on. You might ask yourself whether your supervision question has been answered; whether the supervisor made it easy for you to use supervision; whether the balance of informality and rigidity, being didactic and being non-directive, or support and constructive criticism was right.

Other aspects of your contribution to successful supervision include: keeping on track; being as open as possible, rather than feeling you have to show only competence; and not being defensive about admitting difficulties. You will not learn much from supervision if you say everything is OK and your supervisor says that sounds fine! Everyone does imperfect therapy, and your best chance of getting more skilled is through openness about what goes wrong.

Finally, remember that supervision can involve a range of techniques beyond straightforward verbal discussion of cases. It is useful to use role play, in either of two ways. Either you play the client and the supervisor models how you might respond to that client; or the supervisor plays the client whilst you rehearse a particular strategy as therapist (tapes come in useful here – it is much easier for supervisors to play your clients if they have heard or seen them on tape). There may also be times when straightforward didactic teaching or recommended reading is a useful part of supervision.

Problems with supervision

Not being able to find a supervisor

Although we must stress the importance of securing regular, competent supervision, you might sometimes find yourself between supervisors or unable to contact your supervisor. We would then suggest that you first consider self-supervision. Set aside time to listen to your therapy recordings and to review your supervision question (perhaps using the Padesky guidelines above) and critically reflect on your practice and resources – in short, discover whether you can resolve the difficulty yourself. If you cannot, you could consider trying to obtain a one-off consultation from a colleague or external ‘expert’. Do not simply resign yourself to not having supervision.

Problems with the supervisor

Common problems include a lack of commitment from the supervisor (e.g. regularly being late or interrupting supervision sessions to take phone calls, etc.); or feeling poorly supported by the supervisor. It is possible that such problems arise because a supervisor and supervisee are not well matched and an alternative arrangement is needed. However, it is worth considering if the working relationship can be improved. Is the supervisor aware of your needs? If not, why not? Are your expectations of your supervisor realistic, or do you need to supplement your supervision with further support or personal therapy? Investing time in establishing a mutually agreeable contract can often prevent or minimise problems arising in the supervisory relationship.

Feeling unable to bring along tapes of sessions for evaluation

Try to discover what is handicapping you. Perhaps you need to think through the reasons for using tapes; or perhaps you need to do a BE to see whether the pros or cons are as you imagine them to be.

Is the problem performance anxiety or fears of being ‘found out’ as incompetent? That kind of worry is extremely common, even in experienced therapists, and it is important to try to work with it. Use guided discovery to help you unpack the reason(s) for your reluctance, and consider taking the problem along to supervision.

Agreeing strategies that are not carried out

Again, you need to understand the thoughts and beliefs that account for this. For example, is it over-compliance with your supervisor, agreeing to take actions that you do not really agree with? If so, what makes it difficult for you to tell your supervisor that you disagree?

Or is it that you genuinely think the plans are good, but then in your therapy sessions you get sidetracked? Again, try to understand what leads to that. Therapy tapes may be very helpful in looking at where therapy sessions go off track. And, as always, you can take these questions to supervision.

Negative beliefs about supervision

We have occasionally come across therapists who have negative views about supervision. For instance, the primary purpose of supervision may be seen as ‘overseeing’ or making sure that it is done ‘right’, and it may therefore be perceived as an aversive experience in which you are likely to be subjected to managerial control or trenchant criticism. On review, it might well become apparent that the primary purpose of supervision is usually to help you get better at what you do and to be able to help your clients more effectively. Supervision could therefore be seen as an opportunity to learn, not as a threat. As in our clinical work, it is crucial to identify the negative beliefs that interfere with progress. Thus, you need to be aware of your own unhelpful beliefs and use your CBT skills to evaluate them.




  • Although we need more research evidence, there is an almost universal consensus that adequate clinical supervision is an essential part of becoming a competent CBT therapist (or indeed any other kind of therapist).
  • Supervision can be done in different formats (pairs, groups, etc), via different channels (face-to-face, phone, etc.), and with or without a recognised leader. Although what we have called the ‘apprenticeship’ model is probably the most common, all have their pros and cons, depending on individual circumstances.
  • In CBT, ‘live’ supervision (i.e. the supervisor having direct access to the session, by sitting in or by listening to recordings) is highly valued and is also now required for BABCP accreditation as a therapist.
  • If you have a choice, give some thought to who would be the best supervisor for you, and agree the terms of the supervision contract (whether it is an actual contract or just an informal understanding).
  • Supervision is most useful if you prepare for it, and identify beforehand what your supervision question is (i.e. what you want to get out of the supervision session).

Learning exercises

Review and reflection:


  • Spend a little time reflecting on your current supervision (and perhaps arrange to reflect together with your supervisor as well). What is working well? What not so well? What are the main learning points you have got out of it? Is there anything you and/or your supervisor need to do differently to get more out of it?
  • Are you currently using ‘tapes’ (session recordings)? If not, think about what combination of factors is stopping you: technological difficulties, problems stemming from a lack of confidence, particular beliefs about what might happen, or whatever else it might be.

Taking it forward:


  • Could you make better use of supervision time if you do more preparation? Perhaps you could use the Padesky supervision questions described above as a framework, and see whether it is helpful.
  • If you currently have fears that stop you from playing session recordings to your supervisor, perhaps you could think about doing a behavioural experiment to see how it goes if you try it out? Is it as bad as you feared? What are the actual advantages and disadvantages?
  • If you are starting out with CBT and you do not yet have a CBT supervisor then get one as soon as possible!

Further reading

We are not aware of any other significant texts on receiving supervision, as opposed to doing supervision. The following are therefore suggested as useful references on the general topic of CBT supervision.

Bennett-Levy, J. (2006). Therapist skills: their acquisition and refinement. Behavioural and Cognitive Psychotherapy, 34, 57–78.

An interesting and influential attempt to build a model of how therapists become more skilled and how different aspects of skill may need different learning methods.

Kennerley, H., & Clohessy, S. (2010). Becoming a supervisor. Chapter 16 in M. Mueller, H. Kennerley, F. McManus, & D. Westbrook (Eds.), Oxford guide to surviving as a CBT therapist. Oxford: Oxford University Press.

A guide for new supervisors on the challenges that arise and the skills that are needed for good supervision.

Milne, D. (2009). Evidence-based clinical supervision: principles and practice. Chichester: Wiley–Blackwell.

Derek Milne, one of the leading UK researchers in supervision, provides a summation of the available evidence about supervision, and a guide to best practice based on that evidence.

Padesky, C. (1996). Developing cognitive therapist competency: teaching and supervision models. In P. Salkovskis (Ed.), The frontiers of cognitive therapy. New York: Guilford Press.

In this chapter Padesky argues that CBT supervision can usefully parallel the approach of CBT therapy; and she also outlines the useful ‘supervision checklist’ referred to earlier in this chapter.