What is group CBT?

Group psychotherapy has its origins in psycho-dynamic models of pathology and it developed before the practice of CBT was first outlined by Aaron Beck and colleagues (Reference Beck, Rush and ShawBeck 1979). Importantly, group psychotherapy views the interactions between the group members as the vehicle of change. In CBT groups, it has traditionally been assumed that the cognitive–behavioural model taught to the group is more of an ‘active ingredient’ than the relationships between the group members (Reference Bieling, McCabe and AntonyBieling 2006). Indeed, the educational ethos is inherent in the CBT model, and this lends itself very easily to the provision of groups or classes, as does the fact that CBT is structured and directive, collaborative and time-limited (Reference Fennell, Hawton, Salkovskis and KirkFennell 1989).

However, group CBT cannot solely be about taught skills, and in common with individual (one-to-one) CBT, non-specific interpersonal factors play a part. As with individual CBT, these non-specific factors have not attracted the same research attention as the underlying cognitive–behavioural theoretical models (such as which underlying thoughts and behaviours are paramount in different disorders) (Reference Bieling, McCabe and AntonyBieling 2006; Reference Tucker and OeiTucker 2007). This is in sharp contrast to group psychotherapy, which has focused on interpersonal relationships and related mechanisms of change. The most influential of these suggested mechanisms have been Yalom's ‘curative factors’, listed in Box 1 (Reference YalomYalom 1995).

How does group CBT work?

To varying degrees, Yalom's therapeutic factors are likely to be relevant to all models of group psychotherapy. Wherever you have the formation of a group you will have patterns of relating – ‘group processes’ – that will affect both the group as a whole and the extent to which group members benefit from attendance (Reference Burlingame, Fuhriman, Johnson, Lambert, Bergin and GarfieldBurlingame 2004). Some of Yalom's factors – such as the corrective recapitulation of the primary family group – might be seen as less relevant to CBT groups. However, researchers have translated aspects of the concept into CBT principles – so that the group might be seen to modify maladaptive relational patterns through observing others and trying out new styles of interacting (Reference Bieling, McCabe and AntonyBieling 2006).

White & Freeman describe the two important elements that need to be present for an effective CBT group as the group's cohesiveness and task focus. They define cohesiveness as ‘the degree of personal interest of the members for each other’ (Reference White and FreemanWhite 2000: p. xiii). All CBT groups should be task-focused, in that there should be defined goals to be achieved. Other mechanisms that probably play a part in effecting change in group CBT are listed below: as can be seen, many incorporate concepts inherent in Yalom's therapeutic factors. For a more detailed consideration of the potential mechanisms of change in CBT groups, I recommend Reference Bieling, McCabe and AntonyBieling et al (2006).

How can group CBT be delivered?

Who can benefit from group CBT?

In addition to whether there is an evidence base supporting the use of group CBT for a particular condition or problem, other patient factors need to be considered. Box 2 summarises factors that are believed to influence therapeutic success in individual CBT (Reference MooreyMoorey 1996). The presence of any of these variables need not preclude CBT: a patient may have multiple Axis I diagnoses or significant Axis II problems and still gain great benefit from CBT. It is simply that these factors need to be taken into account when a clinical decision is made about whether to progress with therapy.

These factors also undoubtedly apply to the consideration of a person's suitability for group CBT. Box 3 outlines additional factors that may indicate whether someone will be able to use or to benefit from group-based therapy. These should be considered alongside the general CBT suitability factors (Box 2). Measured consideration of these factors is needed before a decision is taken to include a person in a particular CBT group. For example, the presence of a more severe or chronic problem, or coexisting personality problems, may make someone suitable for a small group in which close monitoring of risk and mental state is feasible. They may not be suitable for a large psychoeducational group that does not offer this potential. Similarly, patients with severe self-harm may still be treated within CBT groups provided that provisions can be made to manage their behaviour. This can be made easier when other group members have the same or similar presenting problems. Indeed, CBT groups are usually populated by people who have the same type of problem or diagnosis. The major reason for this is that the model of CBT used (and associated treatment approaches) varies greatly between disorders, far more than is usually the case for many other psychotherapies (e.g. psychodynamic psychotherapy). Placing people with very disparate diagnoses in a single CBT group is likely to result in treatment that is confusing and at any given point will not be directly relevant to some of the group members.

Other factors may suggest that a candidate for CBT may be able to tolerate group therapy more easily than individual therapy. For example, the patient may feel that individual therapy is ‘too intrusive’, with group therapy allowing for greater anonymity. This will probably make larger psychoeducational groups particularly attractive. Some patients have a history of forming very strong and sometimes destructive (often dependent or regressive) relationships in one-to-one therapy – no matter what the model. Although there is no guarantee that this will not happen in group CBT (towards therapist or other participants), these patients may tolerate a group environment better and make behavioural change without seemingly being derailed.

Is group CBT effective?

There is a substantially larger body of evidence supporting the use of individually delivered CBT compared with the evidence supporting CBT in groups. However, some clinicians have erroneously taken the evidence supporting the former as implicitly supporting the latter. This ignores the likelihood that the format of the therapy, in addition to the underlying model, affects its effectiveness. From the summary of the current evidence base relating to depression and anxiety disorders outlined below, it can be seen that we cannot unequivocally conclude that group and individual CBT interventions have equivalent outcomes. Tucker & Oei have analysed the evidence base for group and individual CBT to calculate which is the most cost-effective (Reference Tucker and OeiTucker 2007). They tentatively deduce that group CBT is more cost-effective for depression, but less cost-effective for anxiety and social phobia. They point out that the available evidence has significant methodological shortcomings (including whether the efficacy trials can be generalised into naturalistic settings) and conclude that ‘It cannot be summarily or definitively stated that group CBT is a cost-effective treatment’.

Conclusions

There appear to be clinical advantages and disadvantages to delivering CBT in group format (Boxes 4 and 5). Some patients appear to benefit most from individualised approaches, whereas others appear to do very well in group CBT. The potential of attracting individuals who do not want the stigma of using formal psychiatric services but who are willing to attend a psychoeducational large group in a community setting such as a leisure centre has real public mental health implications (Reference Brown, Boardman and ElliottBrown 2005). Thus, there is a case for offering a choice of CBT delivery where this can be practically accommodated. There may also be a particular role for group CBT for people who have a very specific problem in common – such as postnatal depression (Reference Milgrom, Negri and GemmillMilgrom 2005) – or for the treatment of depression associated with chronic physical ill health (National Collaborating Centre for Mental Health 2009b). It is worth noting, however, that it is not clear which conditions, and equally which personal characteristics, indicate whether a group or individual approach is likely to be most helpful for a given individual. Questions remain, not only about the relative effectiveness of group CBT compared with individual therapy (Reference Tucker and OeiTucker 2007), but also about their relative acceptability for patients and drop-out rates (Reference Heimberg, Salzman and HoltHeimberg 1993; Reference Sharp, Power and SwansonSharp 2004; Reference Semple, Dunwoody and SullivanSemple 2006).

It has been noted that there is no clear evidence yet for the use of group CBT with people from different cultures (Reference Oei and DingleOei 2008). One critical issue pointed out by Oei & Dingle is that there is currently no coherent theory for group CBT that encapsulates both the content of the CBT model and the interpersonal processes that are occurring in the group. This may go far in explaining why the outcome research is at times conflicting – perhaps it is measuring different processes in different populations. When the theory is clearer we may be able to firmly determine the role of group CBT.

Currently, the NICE guidelines for depression, anxiety disorders (panic disorder and generalised anxiety disorder) and OCD all advocate the use of CBT delivered in groups but only under certain circumstances and within a stepped care framework. For example, they recommend group-delivered traditional CBT only for milder presentations of depression in those who decline low-intensity psychosocial interventions (National Collaborating Centre for Mental Health 2009a). All of these guidelines state or infer that group CBT is either less efficacious than individually delivered CBT or that the evidence base supporting its use is less well developed. Yet the potential of CBT delivered in group format remains – that it may afford the opportunity for patients to learn from each other and for more patients to be treated together with greater efficiency.

Innovative ways of delivering CBT in groups have been developed, as outlined in this article (e.g. the large-group format), but other potential methods are only just being explored. One such is the novel use of intensive CBT groups in the CBT department of Leicestershire Partnership NHS Trust. Patients can elect to be treated over a shorter time frame – perhaps over 4 weeks, but more intensively, with three or four sessions a week rather than the traditional one session weekly. These groups are based on similar intensive CBT developments in individual therapy (Reference Abramowitz, Foa and FranklinAbramowitz 2003; Reference Storch, Merlo and LehmkuhlStorch 2008). Initial results from our department show that this approach provides reasonable outcomes. The intensive format is very popular with many patients for a variety of reasons – most commonly because leave from work and arrangement of childcare is often easier to organise as a block over a short period of time rather than for a few hours a week over a longer period. This is an example of how creative service development can produce potential benefits and choice.

In England, the interventions provided by IAPT services (Department of Health 2008) will be dependent on the relevant NICE recommendations. It is clear that group CBT has a role within these services and despite the often tight stipulations provided by NICE, such a large development of psychological therapy services affords us with the opportunity to further innovate and measure the effectiveness and acceptability of differing approaches, including CBT in group format.

MCQs

Select the single best option for each question stem

1. 1 In group CBT, a cohesive group

   1. a is not believed to predict a better outcome

   2. b refers to how much the CBT model is relevant to the patients' problems

   3. c is likely to be most cohesive at the beginning of therapy

   4. d helps the group focus on the work to be achieved

   5. e is not (by definition) influenced by the therapist.

2. 2 CBT in group format:

   1. a is particularly useful in the treatment of psychotic depression

   2. b has been shown to be a more effective treatment of OCD than has individual CBT

   3. c is recommended within a stepped care model by the updated (2009) NICE guidelines on depression

   4. d is known to be better used by men than by women

   5. e almost certainly leads to better outcomes compared with individual CBT in the treatment of social phobia.

3. 3 Patients who are likely to be potential candidates for group CBT rather than one-to-one CBT include:

   1. a those who have not responded to group CBT in the past

   2. b those with very severe OCD

   3. c those noted to be extremely competitive with their peers

   4. d those with mild depression

   5. e those who express a preference for individual therapy.

4. 4 Large-group CBT approaches:

   1. a usually do not have a psychoeducational emphasis

   2. b characteristically use individualised formulations to dictate all of the content of the intervention

   3. c are used as a treatment for anxiety problems

   4. d cannot be used by patients concurrently receiving other treatments or therapies

   5. e are particularly useful for treating large numbers of people with PTSD in leisure centres.

5. 5 In England, the Improving Access to Psychological Therapies programme:

   1. a aims to increase access to both talking therapies and antidepressant medication

   2. b involves the development mainly of secondary care services

   3. c involves both ‘low’ and ‘high’ intensity treatments

   4. d will focus on bipolar disorder and schizophrenia

   5. e is unlikely to lead to the provision of group CBT.

MCQ answers

1

d

2

c

3

d

4

c

5

c