The aims of therapy for bipolar disorder are to alleviate acute symptoms, restore psychosocial functioning, and prevent relapse and recurrence. The benefits of pharmacotherapy have dominated the literature on treatment. However, the efficacy of prophylaxis in research settings has not been matched by effectiveness in clinical practice. Lithium prophylaxis protects only 25-50% of people with bipolar disorder against further episodes (Reference Dickson and KendallDickson & Kendall, 1986), and the introduction of newer medications has not improved prognosis (Reference ScottScott, 1995a ).

Since the 1980s there has been a greater emphasis on stress-diathesis models. This has led to the development of new aetiological theories of severe mental disorders which emphasise psychological aspects of vulnerability and risk, and has also increased the acceptance of cognitive therapy as an adjunct to medication for individuals with treatment-resistant schizophrenia and with severe and chronic depressive disorders (Reference Scott, Wright, Frances and HalesScott & Wright, 1997). Research in bipolar disorder has been limited, but there is evidence that cognitive therapy may benefit this patient group.

IS THERE A PSYCHOLOGICAL MODEL OF BIPOLAR DISORDER?

At present, psychological theories of bipolar disorder have been insufficiently researched, and there are few insights to aid our understanding and explain how therapy achieves its effects (Reference Lam and WongLam & Wong, 1997; Reference Scott, Stanton and GarlandScott et al, 2000; Reference Palmer, Scott, George and BirchwoodPalmer & Scott, 2001). However, it is possible to use an adaptation of Beck's original cognitive model of depression and mania to increase its applicability to bipolar disorder (Reference Beck, Rush and ShawBeck et al, 1979; Reference ScottScott, 1995b ; Reference Beck and SalkovskisBeck, 1996). The description of this model should not be regarded as an empirical explanation of the aetiology of this disorder, but it provides a framework for understanding its psychopathology, and how cognitive and behavioural approaches may be targeted at specific symptoms and problems. Other psychological models of bipolar disorder, such as the ‘interacting cognitive subsystems’ and the ‘behavioural engagement system’ theories, are also being explored.

IS THERE A CLINICAL RATIONALE FOR USING COGNITIVE THERAPY?

The above description represents an unproved but important theoretical model (Reference Palmer, Scott, George and BirchwoodPalmer & Scott, 2001). While research continues on this and other models, clinicians should remember that there are some obvious and pragmatic reasons why cognitive therapy may benefit patients with bipolar disorder. First, cognitive therapy shares the characteristics of a clinically effective short-term psychological intervention (Reference ScottScott, 1995c ). Specifically, the therapy is highly structured and based on a coherent model, provides the patient with a clear rationale for the interventions made, promotes independent use of the skills learned, attributes change and progress to the individual's rather than the therapist's skilfulness, and enhances the individual's sense of self-efficacy. The second reason why cognitive therapy may be beneficial is that the clinical picture can be formulated into an acceptable generic or specific cognitive therapy model for all patients. At a specific level, understanding bipolar disorder and its impact on the individual requires a conceptualisation that encompasses cognitive (thoughts, images and beliefs), behavioural, effective, biological and environmental areas of the individual's life.

The approach outlined in Fig. 1 (adapted from Reference Basco and RushBasco & Rush, 1995) is particularly useful when working with individuals with bipolar disorder as it allows the therapist to emphasise a stress-diathesis model that may also include biological factors as precipitants of symptom shift. In order to use this approach, the therapist should first ask about the patient's own views on the causes of the disorder and the associated problems. The patient's aetiological theory is then incorporated within the framework of the model. Links between the individual's cognitions, behaviour, mood and other symptoms (particularly sleep disturbance) and the interaction between these and the environment (stressful events or experiences that are a cause or a consequence of other shifts) are emphasised. This rationale is used to engage the patient in cognitive therapy through monitoring and linking changes in thoughts, behaviours, feelings and the biological symptoms of bipolar disorder. The model also acknowledges that sleep disturbance may be a useful predictor of biological and/or psychosocial disruption and may act as an early-warning sign of shifts from euthymic to abnormal mood states (Reference Wehr, Sack and RosenthalWehr et al, 1987). When the connections between the biological and other aspects of their experience are exposed, patients are able to understand the reasons for using cognitive therapy as well as medication. This establishes the rationale for cognitive and behavioural interventions, and also provides a starting point to explore attitudes towards the use of, and adherence to, medication.

Fig. 1

Conceptualisation of bipolar disorder (adapted from Reference Basco and RushBasco & Rush, 1995).

The above approach is generally well received, particularly in the early stages of cognitive therapy, when the patient is trying to make sense of the disorder and the proposed treatment. Virtually all aetiological ‘theories’ can be incorporated in the model outlined, giving the patient confidence that the therapist is listening to these ideas and is trying to help make sense of them.

OVERVIEW OF COGNITIVE THERAPY

The goals of cognitive therapy in patients with bipolar disorder are to facilitate acceptance of the disorder and the need for treatment; to help the individual recognise and manage psychosocial stressors and interpersonal problems; to improve medication adherence; to teach strategies to cope with depression and hypomania; to teach early recognition of relapse symptoms and coping techniques; to improve self-management through homework assignments; and to identify and modify negative automatic thoughts, and underlying maladaptive assumptions and beliefs (see Reference Basco and RushBasco & Rush, 1995; Reference ScottScott, 1995b ). At the first session, patients are encouraged to tell their story and to identify problem areas through the use of a life chart. Current difficulties are then classified under three broad headings, as intrapersonal (e.g. low self-esteem, cognitive processing biases), interpersonal (e.g. lack of social network) and basic problems (e.g. symptom severity, difficulties coping with work). These are explored in 20 therapy sessions, held weekly for about 15 sessions and then with gradually reducing frequency. After this active phase, two ‘booster’ sessions are offered to patients (at about weeks 32 and 38) to review the skills and techniques learned. The overall programme comprises the following approaches (Table 1).

Table 1

Typical cognitive therapy interventions in bipolar disorder

General coping strategies	Relapse prevention
Education and increasing knowledge	Identifying and modifying dysfunctional beliefs
Discussing symptoms, effect on lifestyle	Stressful events (specific personal meaning)
Information about treatments and outcome	High-risk behaviours (alcohol, drug use)
Encouraging questions, providing information	Relapse ‘signature’: identify two or three symptoms that are early-warning signs of relapse (avoid using mood, as misattributed as health)
Self-regulation
Sleep, eating, exercise routines	Identify three components: depression-specific, mania-specific and idiosyncratic
Structured daily diary
	Determine action plans, e.g. increasing self-regulation and self-monitoring
Self-monitoring
Mood, thoughts, behaviours	Identify and modify core beliefs
Changes in symptoms with events or treatment
	Crisis management strategies
Adherence management	Hierarchy of coping: stepwise action plan
Assume patient will not always adhere to medication regimen	Avoidance of major decisions
	Prior agreements for action with family and clinicians
Explore fears, attitudes and thoughts
Discuss advantages and disadvantages
Identify factors that increase or decrease risk of non-adherence

DOES COGNITIVE THERAPY IMPROVE OUTCOME?

Encouraging anecdotal and single case reports on the use of cognitive therapy in patients with bipolar disorder have been published over the past 20 years (Reference Chor, Mercier and HalperChor et al, 1980; Reference Wright, Schrodt, Freeman, Simon and BeutlerWright & Schrodt, 1992; Reference ScottScott, 1995b ). There is also a growing number of reports on the use of individual and group therapy in small-scale open or randomised controlled trials (Reference CochranCochran, 1984; Reference Palmer, Williams and AdamsPalmer et al, 1995; Reference Bauer and McBrideBauer & McBride, 1997; Reference Perry, Tarrier and MorrissPerry et al, 1999; Reference Lam, Jones and BrightLam et al, 2001; Reference Scott, Garland and MoorheadScott et al, 2001).

The aim of Cochran's study was to apply cognitive therapy to enhance adherence to prescribed lithium treatment. It compared 28 patients randomly assigned to either six sessions of group cognitive therapy or standard clinic care (Reference CochranCochran, 1984). Following treatment, enhanced lithium adherence was reported in the therapy group, where only three patients (21%) discontinued medication, compared with the standard clinic care group, where eight patients (57%) discontinued medication. There were fewer hospitalisations in the group receiving cognitive therapy.

In an exploratory study by Palmer et al (Reference Palmer, Williams and Adams1995), six patients with bipolar disorder were offered cognitive therapy in a group format. The focus of the programme was psychoeducational, recognising the process of change, strategies for coping and interpersonal problems. Overall findings indicated that group therapy combined with mood-stabilising medications was effective for some of the participants. Improvements were found in symptomatology and in overall social adjustment. All participants improved on one or more measures, although the pattern of change varied across individuals. A more recent study by Palmer et al (presented at the annual meeting of the British Association of Behavioural and Cognitive Psychotherapists, Durham, UK, 1998; further details available from J.S. upon request) included a larger number of participants receiving group cognitive therapy (n=25) and a comparison group (n=12) receiving only treatment as usual. The results showed that there was no difference in mood state between groups, but, in comparison with the control group, cognitive therapy reduced non-specific symptomatology and increased social adjustment.

The Life Goals programme developed by Bauer & McBride (Reference Bauer and McBride1997) is a structured, manual-based intervention which seeks to improve patients' management skills and their social and occupational functioning. Although outcome data are not yet available, an analysis of data on process goals showed that the manual could be used by therapists inexperienced in treating bipolar disorder. Furthermore, the programme was effective in terms of knowledge gained and goals achieved by patients (Reference Bauer and McBrideBauer & McBride, 1997).

Perry et al (Reference Perry, Tarrier and Morriss1999) undertook the largest study so far (n=68), using cognitive and behavioural techniques to help people identify and manage early-warning signs of relapse in a group of patients at high risk of further episodes of bipolar disorder. In comparison with the control group, the intervention group had significantly fewer manic relapses, significantly fewer days in hospital, higher levels of social functioning and better work performance. However, the intervention did not have a significant impact on depression, and the researchers suggested that more formal cognitive therapy might be important to produce significant changes in this area.

A pilot study by Scott et al (Reference Scott, Garland and Moorhead2001) examined the effect of 20 sessions of cognitive therapy in 42 patients with bipolar disorder. Patients were randomly allocated to this therapy in addition to treatment as usual, or to a waiting-list control condition of treatment as usual for 6 months followed by cognitive therapy plus treatment as usual. Preliminary results show cognitive therapy plus treatment as usual may offer some benefit to about 70% of these patients and is a highly acceptable treatment intervention (90% of subjects endorsed this approach). The results also suggest that cognitive therapy plus treatment as usual v. treatment as usual alone may lead to fewer relapses, as well as reducing levels of both depression and mania, and improving self-esteem and self-reported adherence to medication. An out-patient study of similar size (n=25) and design demonstrated comparable results, with additional significant improvements in social adjustment (Reference Lam, Jones and BrightLam et al, 2001).

CONCLUSION

Cognitive therapy appears to be an effective approach in the treatment of bipolar disorder; but the development of a more coherent and empirical model will require a greater understanding of individual vulnerability factors that may influence onset or outcome of episodes of mania, depression or mixed states. More research on commonly shared underlying beliefs in individuals at risk of this disorder, attitudes towards the disorder and views about its treatment will also aid our understanding of adjustment difficulties and psychological barriers to engagement or adherence to treatment and good outcome.

On the evidence available, it appears that cognitive therapy has particular characteristics that may benefit patients with bipolar disorder. Its collaborative, educational style, and the use of a stepwise approach and of guided discovery, make it acceptable to individuals who wish to take an equal and active role in their therapy (Reference Beck, Rush and ShawBeck et al, 1979; Reference ScottScott, 1995b ). This is important as many individuals with this disorder resist and challenge a didactic approach to treatment (Reference Miklowitz and GoldsteinMiklowitz & Goldstein, 1990). However, cognitive therapy is not the only short-term therapy that can be used: psychotherapies of proven effectiveness in major depressive disorder, such as interpersonal and family therapies, are also being piloted in bipolar disorder.

Finally, larger-scale treatment intervention studies will be needed to establish the effectiveness of cognitive therapy and to differentiate between the specific and non-specific benefits of therapy. Ultimately, it is important for clinicians and researchers to know not only whether the use of this form of therapy is indicated, but also whether it helps the individual with bipolar disorder because it enhances medication adherence, changes cognitive representations of the disorder, leads to the development of compensatory skills, or reduces vulnerability to relapse through schema change. Such research will also allow for the further exploration of the complex interplay between psychological, social and biological factors.

Clinical Implications and Limitations