Category: CBT

Is it possible to predict who will benefit from cognitive behavioural therapy (CBT)?

The rise of CBT has been welcomed by many as safe, effective alternative to drug treatments for mental illness. However, there are also fears that CBT has grown too dominant, crowding out other less structured, more time consuming forms of psychotherapy.

The fact is, CBT doesn’t work for everyone. Precious resources could be better managed, and alternative approaches sensibly considered, if there were a way to predict in advance those patients who are likely to benefit from CBT, and those who are not.

Jesse Renaud and her colleagues administered a ten-item scale – the Suitability for Short-term Cognitive Therapy, first devised in the 90s – to patients who underwent CBT for depression or anxiety at the McGill University Health Centre between 2001 and 2011. The researchers focused their analysis on the 256 patients (88 men) who completed their course of therapy, which lasted an average of 19 sessions.

Renaud’s team looked for correlations between patients’ answers to the Suitability scale and found that the scale was really tapping two main factors – the patients’ capacity for participation in the CBT process, and their attitudes towards CBT.

The first factor includes a patient’s insight into thoughts that pop into their heads (so-called “automatic thoughts”); their ability to identify and distinguish their emotions; and their use of safety behaviours to cope with their problems (e.g. avoiding parties to cope with social anxiety). In other words, the researchers explained, this is the patient’s “ability to identify thoughts and feelings, and share them in a non-defensive, focused way.” The second “attitudes” factor refers to, among other things, the patient’s optimism about the outcome of therapy, and their acceptance that they must take responsibility for change.

The higher patients’ scored on the first factor (their capacity for participation in CBT), the greater reduction they tended to show in their illness symptoms, based on measures taken before and after the course of CBT. Attitudes towards therapy were not correlated with symptom reductions, but we should bear in mind that this may be because the research focused only on those patients who completed therapy. Also, it may be useful in future to measure how patients’ attitudes change during therapy.

There are other reasons for caution. The amount of variance in symptom change explained by both suitability factors combined was statistically significant, but tiny – just .07 per cent. Also, the same therapists who administered the therapy, recorded their patients’ improvements, so there was clearly scope for bias. Finally, more research is needed on different forms of mental illness besides depression and anxiety. Nonetheless, this study makes a constructive contribution to a neglected area.

“Given that the patient’s capacity provides important information about whether or not a patient will derive benefit from CBT, clinicians who are concerned about limited resources and long wait lists are encouraged to undertake a suitability assessment prior to therapy,” the researchers said, “identify patients low in their General Capacity to Participate in the CBT Process, and consider making referrals to alternative treatments (e.g. other psychotherapeutic approaches, pharmacotherapy.”


Renaud J, Russell JJ, & Myhr G (2014). Predicting Who Benefits Most From Cognitive-Behavioral Therapy for Anxiety and Depression. Journal of clinical psychology PMID: 24752934

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

What is cognitive behavioural therapy like for a teenager?

Most research into CBT (cognitive behavioural therapy) for teenagers has focused on whether it works or not, with largely positive results. Surprisingly little attention has been paid to finding out what it is actually like for a teenager to undertake CBT.

Deanna Donnellan and her colleagues have made an initial effort to plug this gap, conducting in-depth interviews with three teenage girls who’d completed a course of individual CBT, asking them about their perception of the therapy and what it meant to them.

The pseudonymous interviewees were Mary, who had problems with sickness and anxiety; Katherine, who had anxieties around her appearance and restricted her eating; and Samantha, who experienced low mood and practised self-harm. The teenagers were aged 15 years on average.

One the main themes to emerge related to progress and change. Mary saw the therapy in terms of helping to remove her problems; Samantha saw it as more than that, as a chance to move forward in her life; and Katherine felt she had developed new perspectives on life and the future. All three experienced increases to their self-efficacy (their confidence in their own abilities). Donnellan and her colleagues pointed out a related practical insight here – they found the teenagers clearly had “ultimate goals” for therapy (such as a growth in character or a return to “normality”), which could be hidden beneath the immediate aims of the CBT.

Another key theme to emerge related to engagement with therapy. The teens were mostly disengaged and passive at the start, but they gradually began to participate more. Mary achieved this engagement by taking some control – she agreed to take on some of her homework tasks around eating, but refused others. Samantha didn’t say much at the start, but came to realise that she could benefit from exploring her emotional issues. Katherine felt desperate and unable to make decisions at the start, but the graded nature of the therapy helped her feel more stable.

The researchers said issues of control were very important in teen therapy given that most teenagers’ therapy will have been instigated by their parents. “Power and its ability to impact negatively upon therapeutic potential might … be mitigated by a process of collaboration and encouraging the client to negotiate their position in the therapeutic relationship,” they said.

What about rapport with the therapist? Although she benefited from therapy, Mary was not on the same page as her therapist:

“for an example she might use someone being scared of dogs and how the thoughts of the dog biting them would make them cross the road (…) it was like relates nowhere near to like feeling sick and how feeling sick affects ya it was nothing near that”.

Mary blamed part of this on her therapist seeming “really old”. “I think for most teenagers,” Mary said, “… you’d feel easier to talk to someone who, not obviously dead young, but d’ya know not someone in their 50s or something or like old.” In contrast, Samantha was pleasantly surprised at her therapist’s ability to relate to her situation:

“It was a bit disconcerting cos she like, not knew about it, but knew how to like deal with all this stuff, which I wasn’t entirely expecting but it was helpful.”

The final theme related to the structure of the way therapy was delivered. Mary felt like some of the progress was too slow and there was frequent repetition. For Samantha, the structure and predictability of CBT was an advantage, and the boundaries laid down by her therapist helped her feel safe. Katherine also liked the graded pace of therapy, with the gentle start helping her to feel more comfortable.

Donnellan’s team said their interviews were a “tentative” first step towards finding out what CBT is like for young people. The findings demonstrate “the importance of the process of therapy, just as much as the content,” they said. Based on this, some practical recommendations include: recognising the importance of the first stages of therapy for engaging with a teenage client; addressing the teen client’s preconceptions about therapy; and finding out the pace and style they’d like the therapy to progress at.

“The service delivering CBT needs to promote the young person as being in control from the outset,” the researchers said, “regardless of who is making the decision to access therapy. This may set the scene for them to develop control over their problems and establish stability in their life.”


Donnellan, D., Murray, C., and Harrison, J. (2012). An investigation into adolescents’ experience of cognitive behavioural therapy within a child and adolescent mental health service. Clinical Child Psychology and Psychiatry, 18 (2), 199-213 DOI: 10.1177/1359104512447032

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

Trainees in Cognitive Behavioural Therapy underestimate their therapeutic skills

For psychotherapists, the research literature can sometimes make for uncomfortable reading. Yes, most people benefit from therapy, but other findings are less welcome, such as that therapeutic outcomes are unrelated to therapist experience, and that therapists tend to overestimate their skills.

A new study of trainee cognitive behavioural therapists bucks this trend. Freda McManus and a her team have found that several dozen trainee CBT therapists tended to underestimate, not overestimate, how good they were at conducting CBT therapy.

Finding out how accurate therapists are at judging their own skills is important because quality control in therapy often relies on therapists seeking out extra help and supervision when they think they need it.

The new data come from 26 trainees enrolled on the Diploma in Cognitive Behavioural Therapy and 38 trainees enrolled on the MSc in Advanced Cognitive Behavioural Therapy – both courses are at the University of Oxford and the Oxford Cognitive Therapy Centre. The Diploma and MSc students submitted two to six video recordings of therapy sessions they’d conducted. They watched these tapes themselves and rated their own performances. These self ratings were then compared against ratings provided by expert supervisors on the training courses.

Overall the trainees tended to underestimate their skills as compared with ratings provided by their supervisors. Splitting the trainees into two groups – more and less competent – it was the more competent trainees who tended to underestimate themselves. The less competent trainees’ self ratings didn’t differ from the ratings they received from supervisors.

“These results are encouraging in suggesting that CBT therapists may be less susceptible to over-estimation of their competence than has been previously reported,” the researchers said, “which is likely to have benefits for the delivery of CBT interventions in routine clinical practice.”

Why would trainees be underestimating their skills? One explanation lies in a concept known as “defensive pessimism” – a way for high performers to ensure they still receive support and remain motivated to improve their standards. Potentially this is a good thing for clients, but the trainees could suffer in terms of job satisfaction and morale.

There are some question marks over the new findings. For example, the rating scale that was used to assess CBT performance (the Cognitive Therapy Scale) is known to be rather unreliable. Also, it’s possible that the supervisors’ ratings were lenient so as not to demoralise their students. A strength of the study is that the participants were not self-selected – they were all obliged to submit their therapy recordings. By contrast, an earlier study that reported over-confidence in CBT therapists was a highly selective sample obtained by inviting participation. It’s possible that sample may have been biased towards particularly over-confident therapists.

ResearchBlogging.orgMcManus, F., Rakovshik, S., Kennerley, H., Fennell, M., & Westbrook, D. (2011). An investigation of the accuracy of therapists’ self-assessment of cognitive-behaviour therapy skills. British Journal of Clinical Psychology DOI: 10.1111/j.2044-8260.2011.02028.x

Post written by Christian Jarrett for the BPS Research Digest.

Emily Holmes: My inner CBT therapist

Imagine you are about to give the “best-(wo)man’s” speech at your friend’s wedding: vast audience, huge hall, microphone, lights, wine, flowers, expectant faces … but words fail you. Worse than that, I was consumed with an overwhelming feeling of nausea. I’d just found out I was pregnant and not told the world yet. I could see myself about to vomit at the photographer and over the bridal couple … looming panic. “I’ve given hundreds of speeches, it’ll be fine”. But reassuring words alone didn’t help. More nausea. “OK! Stop the internal focus” – my inner CBT therapist suddenly kicked in. “This isn’t real – this is just an image of vomiting”. The inspirational CBT work on mental imagery and social anxiety (David Clark, Ann Hackmann, Colette Hirsch and others) zoomed in. “Focus outwards! Look at the audience “. OK, deploy “cognitive science” – “external perception will compete for resources with internal images. Focus on the flowers”. Oh, and a bit of image restructuring – “mentally photoshop that image of myself, I’m not looking nauseous at all, just moved by emotion at the happy couple”. Here we go … External reality started to win. I was smiling and dinner was staying down. “Good evening everyone …”

Emily Holmes is Professor in Clinical Psychology and Wellcome Trust Clinical Fellow at the
Department of Psychiatry, University of Oxford. She is co-author of the Oxford Guide to Imagery in Cognitive Therapy.

Return to main menu for Psychology to the Rescue

What clients think CBT will be like and how it really is

Some people expect cognitive behavioural therapy (CBT) to be more prescriptive than it is, and therapists to be more controlling than they really are. That’s according to a series of interviews with 18 clients who undertook 8 sessions (14 hours) of CBT to help with their diagnosis of generalised anxiety disorder.

Henny Westra and colleagues selected for interview nine clients whose therapy had ended positively and nine whose therapy had ended poorly. Four of the clients were male. There were four CBT therapists – two men and two women. One was PhD qualified, two were senior clinical psychology grad students, one was junior.

The vast majority of client comments (84 per cent) relating to expectations were that the CBT was not what they’d anticipated. Clients whose outcome was good tended to say they’d been pleasantly surprised – the therapist was collaborative and non-judgmental, and they’d had the opportunity to direct the therapy and choose what to talk about. Of the therapeutic process, the positive outcome clients felt, to their surprise, that they could trust the process, felt comfortable, and that they learned more than they expected. Both good and poor outcome clients worked harder in therapy than they anticipated.

Unsurprisingly, the poor outcome clients tended to say they’d been disappointed by the therapeutic process. In the majority of cases, they took pains not to blame their therapist, instead attributing their lack of progress to time constraints, poor health, their own unrealistic expectations, or their failure to remember the techniques. Direct criticism of the therapist was rare (even though interviewees were reassured their comments were confidential). One person said it would have been better not to have waited until session seven to discuss a key subject from their past.

Sixteen per cent of expectation-related comments conveyed that therapy was just as had been expected. One good outcome client in this category said they thought the therapist would get to the root of their problems, and he did. Poor outcome clients, by contrast, tended to make superficial remarks: ‘it was fairly similar to what I expected, I guess’.

The broader context for this research is that client expectations are one of several factors that are known to be associated with therapeutic success (with positive expectations tending to precede good outcomes). However, very little research until now has looked at expectancy violations – that is, when therapy isn’t what was expected, for good or bad.

‘The findings … suggest that expectancy disconfirmation in CBT, particularly negative expectations for the therapist and the therapy process, is a common and potentially powerful phenomenon in the experiences of CBT clients with good outcomes,’ the researchers said.

A major shortcoming of this research is that the interviews weren’t conducted until after the final therapy session, so it’s possible that clients recalled their earlier expectations in light of their positive or negative experiences in therapy.

ResearchBlogging.orgWestra, H., Aviram, A., Barnes, M., & Angus, L. (2010). Therapy was not what I expected: A preliminary qualitative analysis of concordance between client expectations and experience of cognitive-behavioural therapy. Psychotherapy Research, 20 (4), 436-446 DOI: 10.1080/10503301003657395

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

CBT-based self-help books can do more harm than good

Self-help books based on the traditional principles of CBT, including popular titles like ‘CBT for Dummies’, can do more harm than good, according to a new study. The risks were highest for readers described as ‘high ruminators’ – those who spend time mulling over the likely causes and consequence of their negative moods.

The new research focuses on the use of self-help books as a preventative intervention for people at risk of developing depression. Gerald Haeffel identified 72 undergrads at risk and allocated each of them randomly to work through one of three self-help books. A third of the students spent four weeks working through a traditional self-help CBT-based book, of the kind typically found in book stores, which involved learning the links between thoughts, behaviour and mood, as well as identifying negative thoughts and re-evaluating them. Another group of students followed a ‘non-traditional’ CBT-based self-help book, similar to the first but modified so that the task of identifying and challenging one’s own negative thoughts was removed. The final group followed a book that taught academic skills such as time-management and memory aids.

Here’s the bottom line: among students who tended to ruminate and who had suffered an increase in stress, those who followed the traditional CBT book displayed more depressive symptoms after the four-week study period than those who followed either of the other two books. At four-month follow-up, the traditional CBT study group as a whole tended to have more depression symptoms than the other groups, although high ruminating and stressed students in the traditional group remained the biggest losers.

Haeffel sounded some notes of caution – the findings may not generalise to non-student participants, the samples were fairly small, and the outcomes were based on depression symptoms, not clinically diagnosed depression. That said, the stressed, high ruminators in the traditional CBT group ended up scoring on the ‘moderate’ range of the depression scale at four-month follow up.

‘The current results suggest that cognitive work-books as traditionally operationalised (and sold in stores) may not work for individuals who ruminate,’ Haeffel said. ‘For these individuals, a modified form of cognitive skills training that does not rely on identifying and disputing negative cognitions may be more effective.’

This latest warning about self-help comes after a study published in 2009 that showed use of positive mantras such as ‘I’m a lovable person’ can actually be harmful to people with low self-esteem.

ResearchBlogging.orgHaeffel, G. (2010). When self-help is no help: Traditional cognitive skills training does not prevent depressive symptoms in people who ruminate. Behaviour Research and Therapy, 48 (2), 152-157 DOI: 10.1016/j.brat.2009.09.016

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

Turning talking therapies into doing therapies

There’s plenty of research evidence for the effectiveness of cognitive behavioural therapy (CBT), but for some reason it doesn’t always seem to work so well in real-life settings. In what many psychologists will surely find a readable and helpful paper, CBT expert Glenn Waller outlines why this is often the case, providing solutions along the way.

The biggest single problem, according to Waller, is that real life clinicians often fail to deliver proper CBT with all its active ingredients. For example, one of the most important aspects of CBT is behavioural change, yet clinicians often shy away from encouraging clients to adopt the changes they need to make, especially when such changes are likely to provoke increased anxiety in the short term.

“Many clinicians make the effort to reduce or to avoid immediate patient distress (and hence their own anxiety about whether they are doing the right thing) by being ‘nice’ to the patient,” Waller explained. “However, this short-term strategy means that we do not press for critical therapy tasks to be done, thus leading to long-term therapeutic immobility”.

Waller reminds clinicians they need to work with their clients to agree on a formulation, and an agreed plan of action, with both parties recognising that this plan, while bringing long-term benefit, might well be difficult in the short term.

Clinicians also need to be assertive in bringing structure to sessions. Many clients may well arrive at therapy sessions keen to discuss immediate crises in their life – but spending each session reacting to these crises rather than working through the long-term goals of CBT will prevent any progress being made.

It’s a similar story with so-called “therapy interfering” behaviours – many clinicians find themselves complicit in a client’s avoidance of homework even though this is a crucial part of CBT. Clinicians should remind clients of the rationale for the home-work and how vital it is for lasting change to be achieved.

Waller says one reason clinicians will often avoid challenging their client’s avoidant behaviours, especially if this is stressful for the client, is because they fear being negatively judged. “The clinician needs to know that he or she is not being judged by short-term, necessary negative transitions (e.g. increased patient anxiety),” Waller said. “Rather he or she needs to be helped in supervision to focus on the value of long-term outcomes that probably depend on those negative short-term steps.”

Other advice in the paper includes recognising when therapy isn’t working and bringing it to an end, and resisting switching, without a clear rationale, to so-called “third-wave” therapies, such as schema therapy, which often lack a behavioural change element.

ResearchBlogging.orgWaller, G. (2009). Evidence-based treatment and therapist drift. Behaviour Research and Therapy, 47 (2), 119-127 DOI: 10.1016/j.brat.2008.10.018

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

When clients in therapy show sudden, dramatic improvements

There’s growing evidence that people who undergo psychological therapy often demonstrate sudden, dramatic improvements, almost as though they’ve had a revelatory change of outlook and thinking style. What’s more, these sudden changes appear to be clinically meaningful. People who exhibit sudden improvements from one session to the next are more likely than other clients to show greater and more sustained improvement after they’ve stopped participating in therapy.

Now Elise Clerkin and colleagues at the University of Virginia have investigated the significance of sudden gains among 30 clients undertaking 12 weeks of group Cognitive Behavioural Therapy (CBT) for panic disorder – a context in which the sudden-improvement phenomenon has yet to be studied.

Clerkin’s team found that 43 per cent of clients exhibited at least one dramatic burst of improvement during the course of therapy. Approximately half of these clients showed this improvement between the first and second sessions, while the other half showed their gains later on.

The timing of the sudden improvement proved to be significant. Only those clients who showed dramatic gains after the second session or later tended to show better symptom outcomes at the end of the course of therapy relative to non-dramatic improvers. This makes sense given that the first session was really just an introduction and didn’t include any of the active ingredients of CBT.

Moreover, the later dramatic improvers showed a greater reduction in their fear of anxiety-related symptoms (e.g. a racing heart-beat) at the end of the course of therapy (and at six months’ follow-up) than did the very early dramatic improvers. This suggests that when a dramatic improvement occurred after the second session or later it probably had to do with the clients changing how they interpreted their anxiety symptoms – one of the key goals of CBT. By contrast, very early dramatic improvement may have reflected a meaningless fluctuation of symptoms.

The researchers said more work is needed to find out what psychological processes underlie the effects of a dramatic improvement during therapy. “We suspect these effects occur because of changes in self-efficacy that follow a large, dramatic improvement, which likely engenders hope for further recovery, and enhances commitment to the therapy,” they surmised. “In fact, the sudden gain itself may confer a critical belief change regarding the patient’s ability to overcome symptoms of panic.”

ResearchBlogging.orgE CLERKIN, B TEACHMAN, S SMITHJANIK (2008). Sudden gains in group cognitive-behavioral therapy for panic disorder. Behaviour Research and Therapy DOI: 10.1016/j.brat.2008.08.002

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

How do people change during psychotherapy – the clients’ perspective

Presumably the goal of psychotherapy is some kind of psychological change for the better, but what is that change and how does it happen? Psychological models refer to such things as ‘stages of change’ and assimilation, but few researchers have sought the views of clients who have undergone therapy.

Tim Carey and colleagues conducted loosely structured, hour-long interviews with 18 women and 9 men who had completed an average of six sessions of cognitive-based psychotherapy (either Method of Levels or CBT) for conditions like depression, anxiety or addiction.

The 22 participants who said they had changed during therapy were unable to come up with a definition of psychological change, but they described their experience in terms of acceptance, behavioural changes, new beginnings and a return to positive emotional states.

Accounts of when change occurred tended to be paradoxical – the participants talked of a gradual process that occurred at an identifiable moment. “It was gradual but the realisation was sudden,’ one client said.

Many of the participants could remember the exact moment they became aware a change had occurred: “I could actually hear it,” one participant said; others spoke of their surroundings: “I was in the pool with my husband.”

The clients’ descriptions of how change occurred fell into six themes: motivation and readiness (“I was desperate to get back to my old self”); tools and strategies (“It’s the changes in behaviour that I learned”); learning (“I would take a lot of stuff home to read about assertiveness”); interaction with therapist (“…they don’t judge your character or think they know you”); perceived aspects of self (“I am a strong person mentally”); and the relief of talking (“Let me get everything out, let me relieve myself of everything”).

The researchers said that while many of these insights are not new – for example they point to factors identified as crucial by psychologists like the importance of the therapeutic alliance and readiness to change – what is new is that “these descriptions have come from the people experiencing the change rather than other sources, and the descriptions were not guided by assumptions about any particular stages of change model.”

Carey, T.A., Carey, M., Stalker, K., Mullan, R.J., Murray, L.K. & Spratt, M.B. (2007). Psychological change from the inside looking out: A qualitative investigation. Counselling and Psychotherapy Research, 7, 178-187.

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

Self-help book better than group-CBT for teenagers at risk of depression

These days, cognitive behavioural therapy (CBT) seems to be the psychological treatment of choice for all manner of mental disorders. But according to a new study, when it comes to preventing depression in teenagers, a self-help book might actually be more effective.

Eric Stice and colleagues recruited 225 adolescent school pupils at risk of depression. These teenagers reported experiencing sadness and had raised scores on a measure of depression, but they weren’t actually depressed.

Some of the teenagers then took part in four sessions of group CBT, while others participated in supportive-expressive group therapy (a forum for discussing feelings in a safe environment), expressive writing sessions or diary writing. The remaining students either received ‘bibliotherapy’ in the form of a self-help book called ‘Feeling Good’, or they acted as ‘waiting list’ controls and received no intervention at all.

On the one hand, CBT outperformed most of the other treatments – its benefit versus no treatment was still apparent at two months follow-up, whereas the benefit of supportive-expressive therapy, expressive writing and diary writing only lasted one month.

But on the other hand, it was only the students given the 1980 edition of the book ‘Feeling Good’ who continued to show reduced depressive symptoms at six-month follow up. “The findings have public health implications”, the researchers said “because interventions such as bibliotherapy are very inexpensive and easy to disseminate relative to CBT and supportive-expressive interventions, which require skilled therapists”.

Moreover, drop out was greatest among the CBT teenagers, while being lowest among the teenagers engaged in supportive-expressive sessions or expressive writing, with bibliotherapy drop out being intermediate. “The finding that drop out rates were lowest for two interventions that focussed on emotional expression suggests that these types of programmes are perceived by participants to be particularly worthwhile”, the researchers said.

Stice, E., Burton, E., Bearman, S.K. & Rohde, P. (2007). Randomised trial of a brief depression prevention programme: An elusive search for a psychosocial placebo control condition. Behaviour Research and Therapy, 45, 863-876.

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.