Cognitive-behavioral therapy (CBT) has been shown to be an effective, short-term psychotherapeutic intervention for the treatment of unipolar depression. Furthermore, in recent years the applications of CBT have expanded to the treatment of personality disorders, eating disorders, and various anxiety disorders. As well, several studies have shown promising results using CBT for the treatment of bipolar disorder (BD). Although there are various psychotherapy modalities that are currently being empirically tested, there are several unique aspects to CBT that make it particularly suitable to the treatment of BD. First, the psychoeducational nature of CBT, which promotes, monitoring and self-regulation, makes it helpful for the treatment of this severe chronic recurring disorder. Second, CBT has been shown to be effective in increasing compliance to pharmacological treatment. This is particularly useful as non-compliance to pharmacological treatment has been reported to be over 50% in treatment, as usual. Third, the established effectiveness of CBT in the prevention of relapse in unipolar depression suggests that it might be useful for relapse prevention in patients diagnosed with BD (NIMH Collaborative Study reported 70% of patients experiencing multiple relapses over a 5-year period). Fourth, preliminary results indicate that the interaction of cognitive style and stressful life events may predict depressive symptomatology in BD. This would also support CBT for the treatment of BD as cognitive styles between unipolar disorder and BD may be similar.
Outcome and psychosocial functioning in BD
Before discussing the application of CBT to the treatment of BD, it is noteworthy to mention that CBT for BD has developed in the context of substantial available data showing the poor outcome of this illness, despite aggressive pharmacological treatment, in about 50% of patients. BD, which affects 1.2% of the population, is a common disorder causing substantial morbidity and mortality. The burden of suffering to the patient and family is enormous and up to 15% of patients with this disorder commit suicide. For many years the course of BD was thought to be episodic rather than chronic and deteriorating (e.g. schizophrenia). However, more recent data has challenged the previous reports of good outcome in BD. It appears that despite aggressive pharmacotherapy, which is often successful in treating the acute phase of the illness, many patients continue to experience multiple relapses, subsyndromal symptoms and significant psychosocial deficits. As early as in the pre-lithium era, studies examining chronicity in BD in a total of almost 2500 patients followed over at least 12 years found chronicity rates to range between 0.8% and 56%. The broad range of chronicity rates can be attributed to the heterogeneity of patients and variability in diagnosis and treatment. With the subsequent widespread use of lithium several investigators have re-examined the course of illness in BD. The Iowa 500 study of Tsuang et al. found that 25% of patients with BD had a poor outcome. Several recent well-controlled studies have also examined the course of illness in BD. The NIMH Collaborative Program on the Psychobiology of Depression Clinical Studies followed 53 patients (diagnosed with BD Type I) naturalistically for 5 years and found 18.9% of patients had at least one relapse, 69.8% experiencing multiple relapses, and 11% did not recover over the course of the study. Coryell et al. followed 29 patients with BD for up to 5 years. Results showed that 25% had not recovered at 5 years.
Likewise, Keller et al. found chronicity rates to be different based on the polarity of the index episode in 155 patients diagnosed with BD (ranging from 7% in mania to 32% in mixed states). Several studies have reported on the poor work adjustment, marital functioning and family relationships in patients with BD. Goldberg et al. followed 55 patients with BD and 49 patients with unipolar depression for up to 4.5 years post-hospitalization and found that only 41% of patients had a good overall outcome at 4.5 years. The bipolar group showed more severe work impairment than the unipolar group and more than half of the BD patients were hospitalized at least once during the 4.5 years.
In addition, contrary to earlier reports, it appears that inter-episode recovery may be less than ideal with subsyndromal symptoms remaining clinically significant. Indeed the presence of subsyndromal symptoms as reported by Keller et al. predicted a much higher relapse rate at 2 years, regardless of whether lithium levels were optimal. Furthermore, as reported by Bauer et al., patients with BD often experience a gap between symptomatic and functional recovery during maintenance pharmacological treatment. This points again to the ongoing psychosocial deficits often experienced by these patients following treatment of the acute phase of illness. Finally, Robb et al. reported quality of life in euthymic BD patients to be as severely impaired as patients with current major depression and worse than some chronic medical conditions.
In summary, the substantial data on the poor outcome in more than half of the patients with BD, the presence of significant subsyndromal symptoms persisting beyond the acute phase of illness and the persistence of significant psychosocial deficits have served as the impetus for the development of adjunctive psychosocial treatments in this disorder. Of note, the NIMH Workshop Report on Treatment of Bipolar Disorder recommended in 1990, further research on psychosocial treatments for BD. The goals of psychotherapy and specifically CBT would be to: 1) improve quality of life; 2) decrease the number and:or severity of episodes. Indeed Post (kindling hypothesis of mood disorders), suggests that early intervention in the course of illness is critical to ‘prevent spontaneous episodes and refractoriness to drug treatment’; 3) increase adherence to pharmacological treatment and address the complications of multiple drug regimens; and 4) address gaps in functioning that have resulted from the early onset of this illness.
Psychosocial treatments for BD
Prior to reviewing CBT interventions for BD a summary of other recently developed psychosocial interventions will be presented. These types of interventions have focused on the family, marital functioning or the individual affected by the illness. As well, several psychoeducational approaches, in families or groups, in combination with medications, have been reported. None were controlled except for the Peet and Harvey study, which showed improved compliance and attitude towards lithium in patients receiving the experimental treatment which included educational lectures and a written transcript. Family focused treatment (FFT) for BD, a manualized intervention developed by Miklowitz and colleagues is an adaptation of family focused interventions for schizophrenia. In this manualized intervention, the affected individual is treated in the context of the family. The hypothesis is that the affected individual is influenced by the stress in his/her family. FFT is a combination of psychoeducation, communication skills and behavioral interventions delivered in a structured fashion, over 21 sessions titrated over 9 months. Preliminary unpublished results of a 12-month outcome randomized clinical trials (RCT) of FFT for 101 patients diagnosed with BD showed that patients assigned to the FFT interventions had significantly fewer relapses and longer well periods before relapse than patients assigned to crisis management. FFT also showed significant improvement in depressive symptoms.
The marital intervention developed by Clarkin and colleagues is also a manualized intervention, with 25 sessions titrated over 11 months for spouses of affected individuals. The sessions are psychoeducational marital interventions. Results in 33 spouses randomly assigned to the marital intervention or medication management alone, (mean scores in the experimental group Global Assessment of Functioning (GAF) 64.38–73.00 at 11 months, and in the control group of medications alone, 64.67–65.71 at 11 months) showed an overall improvement in social adjustment and functioning on the GAF scale without any improvement in symptoms. Finally, Frank and colleagues developed the Interpersonal Psychotherapy and Social Rhythm Therapy (IPRST) which incorporates Interpersonal Psychotherapy (IPT) interventions developed for the treatment of unipolar depression (unresolved grief, interpersonal dispute and role transition) with behavioral interventions. Results of a controlled trial of IPSRT versus medication alone in 48 patients treated over a 52-week period showed an increase in the stability of social rhythms (regularization of life styles) in the IPSRT group. This regularization of life style could not be attributed to an improvement in symptoms but seemed to be the result of a specific intervention aimed at regulating daily life styles.
To date there have been seven studies reporting on the efficacy or effectiveness of CBT in BD. Several of these studies are still in progress. Preliminary reports are encouraging and support the feasibility of CBT in BD. The first published report on CBT for BD examined 28 bipolar outpatients maintained on lithium. This was a 6-week trial of brief individual cognitive therapy. Follow-up measures at post-intervention, 3 and 6 months showed a significant improvement in lithium compliance (eight non-compliant patients in the treatment group versus 17 non-compliant patients in the control group) and fewer hospitalizations in the experimental group (two in the experimental group and eight in the control group). Subjects in the psychotherapy group were also less likely to terminate treatment against medical advice (six in the CBT group – 14 in the control group). The second report published by Palmer et al. used a one sample repeated measure design to evaluate the effectiveness of a 17-week group CBT in six patients diagnosed with BD on maintenance mood stabilizers.
Results showed significant improvement for two patients and a trend for the third on the Well Being Scale of the Internal State Scale. They also found overall social adjustment as measured by the Social Adjustment Scale improved from pre- to post-test in these patients. In another study Palmer and colleagues (personal communication) using series of single-case designs and non-equivalent group designs found that 25 patients diagnosed with BD who participated in group CBT experienced a significant improvement in social functioning and a slight reduction in subsyndromal symptomatology. At the recent Third International Bipolar Conference held in Pittsburgh in June 1999, several findings on CBT for BD were presented. Three of these studies reporting on a total of 89 patients used a control group. The first study by George showed a marked difference, using a brief group CBT intervention (nine sessions) in the experimental group with patients identifying early symptom changes related to relapse and initiating contact for treatment. The second study by Hirshfeld-Becker et al.(also presented at the 2nd International Conference on Bipolar Disorder), used a brief group CBT intervention (11 weeks) and found significantly fewer new episodes in the treatment group than in the control group. Rates of euthymia were significantly higher in the CBT group, at baseline post-group and 3 months (respectively, 35%, 65% and 77% for the experimental group and 36% and 33% for the control group). The third randomized trial by Lam et al. reported on a 16-week group CBT intervention in significantly ill patients who had experienced two episodes in the last 2 years or three episodes in the last 5 years. The psychotherapy group experienced significantly fewer bipolar episodes, higher social functioning and better coping strategies for early symptoms. The treatment group also reported fewer subsyndromal mood symptoms. Finally, Zaretsty et al. found that in a matched-case control design currently depressed bipolar patients improved as much as matched unipolar depressed patients on depression measures after 20 sessions of individual CBT.
An interesting finding was that scores on the Dysfunctional Attitude Score (DAS) did not decrease significantly following CBT unlike the unipolar control subjects. All these studies except for the one by Zaretsky et al. (depressed state) involved patients in a euthymic state or with subsyndromal symptoms. This would imply that CBT is an intervention better suited to the ‘non-acute’ phase of illness. Patients in an acute phase of illness, may not be as receptive to CBT, as they may be too ill to process the information necessary to acquire the CBT skills.
Application of CBT to BD
In terms of applications of CBT to BD, the treatment protocol is divided into three phases which address various issues affecting patients’ lives. The first and second phase of the protocol are usually done weekly, each session lasting 2 h. The third phase (which is optional) takes place every 2 weeks over a 3-month period. Patients should have reached a stable (non-acute) phase of illness prior to beginning treatment. Once the acute phase of illness has stabilized, patients often continue experiencing subsyndromal symptoms that can be targeted by CBT. Indeed, one of the explicit goals of CBT is for the patient to develop skills and techniques to reduce these subsyndromal symptoms. Should a patient destabilize and require hospitalization, they can continue to attend treatment depending on the severity of their symptoms.
The first phase of treatment focuses on educating the patient about medications, symptoms and cause of illness. This phase, which should ideally involve family members, provides the affected individual with the necessary information to better manage their illness. It empowers patients with the expertise that will increase their confidence and allow them to eventually take better control of their lives. It also provides the patient and therapist(s) with knowledge that will be useful in challenging many of the dysfunctional beliefs that will need to be changed. Examples of these beliefs include ‘Having BD is a weakness versus an illness’, ‘There is nothing I can do to fight this illness’. Another goal is to educate patients about early symptoms of relapse and develop strategies to prevent a full-blown episode. In the second phase, or skill-training phase, the cognitive-behavioral interventions similar to the ones developed by Beck et al. for unipolar depression are reviewed to address the depressive phase of the illness. The interventions include the specific cognitive and behavioral techniques for depression, contacting their physician to reassess medications, and problem solving around stressful life events that may contribute to the relapse. Interventions described by Rush and Meichenbaum and Turk developed to increase adherence to pharmacotherapy are also introduced. These interventions address many of the dysfunctional beliefs experienced by patients regarding the need to take medications.
Examples of these beliefs include ‘These medications are addictive’, ‘Once I am better there is no need for continuing these medications’. The self-monitoring characteristics of CBT are especially suited to the manic phase of illness, teaching the patient to recognize early manic symptoms, proceed with pharmacological treatment, and implement behavioral intervention that could minimize the impact of the mania. Some examples of these interventions include establishing contracts during a phase of mood stability (i.e. give credit cards or car keys, remove alcohol), to be taken during an acute manic episode. Other interventions include rhythm regulations or decreasing stimulation. Furthermore, the cognitive interventions developed for depression have been modified to address the positively biased thinking often observed in mania. Some of the cognitive errors in mania include an exaggerated positive bias (e.g. ‘I can do it all’, ‘Things will work out no matter what’).
Once the psychoeducation phase and skill-training phase are completed, patients can enter a third phase of treatment whereby they address some of the very specific interpersonal problems and personal problems resulting from this illness. This is the phase of treatment where cognitive restructuring interventions address core beliefs or schemas. One of the goals is to help the ill person to rebuild a more solid sense of ‘self’. Indeed individuals with BD frequently report being ‘lost’ or not knowing who they are. In fact, recent data showing the onset of BD to be earlier than previously thought would support the notion that early onset of the illness impacts on critical phases of development resulting in marked deficits in self-esteem or identity. Likewise, it appears that the impact of hypomania and depression at an early age are significant precisely because they dramatically affect sense of self and postpone important developmental milestones, such as educational achievements, early work experience and important interpersonal relationships. Indeed, for individuals affected early in their lives or for those affected later with severe and disruptive episodes, dysfunctional core beliefs will likely become self perpetuating. Individuals affected at an early stage may already have dysfunctional beliefs that will worsen with the course of illness. Alternatively, another hypothesis is that individuals develop new dysfunctional core beliefs secondary to the effect of BD on their achievements and interpersonal relationships, regardless of the age of onset of the disorder. Examples of these beliefs include a disturbed sense of autonomy, or personal capability, vulnerability to harm or illness and a sense of defectiveness and unlovability. The recognition of maladaptive core beliefs that may have been established by the early onset of the illness and:or traumatic events are important to address as it will help these individuals understand and cope with the specific psychosocial deficits experienced later in their life course. For example, depression occurring during adolescence will contribute to dysfunctional beliefs in the relationship and achievement domains. The belief that one is ‘unlovable’ or not ‘good enough’ will affect the adult in his or her ability to develop and maintain healthy relationships or achieve their full potential. Therefore, difficulties in the interpersonal domain may be the result of early dysfunctional core beliefs and not the result of the most recent episodes.
Another common clinical presentation illustrating the need to address belief processes, is the sense of shame, embarrassment and stigma attached to this illness. Most affected individuals will eventually adjust to the illness. However, individuals affected early and:or severely typically have a strong sense of shame and embarrassment and will have a more difficult time adjusting to BD. Likewise, the frequency of episodes of both mania and depression will interfere with the development of a proper sense of identity. Manic symptoms that are uncontrolled will contribute to an inflated sense of competence that can not be maintained over time, confusing the person as to his or her true abilities. A common question asked by these patients is ‘How do I know whether it is me or the illness that accomplished all these things?’. These unanswerable questions greatly contribute to a sense of incompetence and failure which last beyond the acute episode. These issues will need to be addressed in psychotherapy, and as early as possible in the hope of preventing the establishment of a chronic pattern of illness.
A slightly different process may occur in patients with later illness onset. Such patients may have functioned adequately (not necessarily well) until their first episode (though it may be unclear if symptoms of hypomania or dysphoria were present). These individuals will frequently present with decreased self-esteem and confidence once the acute phase of illness has resolved. This is often in sharp contrast to their perception of functioning prior to the diagnosis of BD. They will frequently report a higher level of functioning that can not be confirmed by family or friends, suggesting an over-idealization of functioning prior to the onset of illness. Indeed these individuals will report how ‘great’ they were doing and how wonderful everything was prior to the illness and how they feel unable to ever meet that level of functioning again. Clinically, it is often difficult to determine whether this ‘greatness’ was partly mood state dependent emerging during the states of hypomania or whether these individuals were truly functioning well. This ‘greatness’ will be missed in the recovery phase and often leads patients to experiencing a profound sense of loss. Therefore, besides the need to address the subsyndromal symptoms, impaired psychosocial functioning, dysfunctional interpersonal relationships, and poor self-esteem, the clinician needs to help the patient re-establish a sense of identity that takes into account the effects of the illness but focuses on the person’s positive capabilities as well.
The article above contains foundational information. Articles below contain optional updates.
- Patelis-Siotis I, Bipolar Disorders, 2001 Feb; Vol. 3
Reflection Exercise #7
The preceding section contained information about cognitive-behavioral therapy: applications for the management of bipolar disorder. Write three case study examples
regarding how you might use the content of this section in your practice.
What are four unique aspects to CBT that make it particularly suitable to the treatment of BD? Record the letter of the correct answer the