CBT for Bipolar Depression
Given the fact that individuals with bipolar disorder spend a large component of their time in a depressed mood state, as well as the fact that bipolar depression accounts for most of the psychosocial disability within bipolar illness, a rational strategy would be to specifically target bipolar depression with CBT, a psychological treatment with strong evidence for efficacy in unipolar depression (including moderate to severe depression).'"''^' Zaretsky et al^'^ compared 11 outpatients with bipolar disorder I and II who were receiving mood stabilizer medication and suffering moderate to acute bipolar depression with a matched group of 11 subjects with recurrent unipolar depression who were receiving no medication and who were experiencing moderate depression. The group with bipolar depression received CBT specifically designed for bipolar disorder, whereas the group with recurrent unipolar depression received standard CBT. The latter group served as control subjects in a quasi-case-control design. At the end of treatment, subjects with bipolar depression improved to the same extent as the subjects with unipolar depression on both subjective and objective measures, however, dysfunctional attitudes still remained elevated in the group with bipolar depression, despite symptom remission. This finding underlines the importance of addressing recovery rather than symptom improvement and remission and, possibly, also argues for the development of a more individualized and comprehensive treatment approach that addresses underlying personality issues, as articulated in the recent bipolar disorder cognitive therapy manual by Newman et al.'' This study is promising because it holds out the promise that CBT could be offered as an alternative to antidepressant medication for bipolar depression, thereby reducing the risk of cycle acceleration."*^'^^ Nevertheless, replication in larger RCTs is necessary before standard treatment guidelines for bipolar depression can be altered.
In an open naturalistic clinical trial, Fava et al'*'' applied CBT to 15 patients with bipolar I disorder who relapsed while receiving effective lithium prophylaxis, despite initial response and adequate medication adherence. The 10 individual CBT sessions of 30 minutes each were conducted by experienced therapists and focused on treating the residual symptoms of bipolar disorder, lifestyle modification, and PE. Mean follow-up ranged from 24 to 108 months. Of the 15 patients, 5 relapsed during follow-up (4 within 30 months); mean survival after CBT was 76 months (standard deviation 13 months). There was a significantly longer time to relapse in the subjects after CBT, compared with before CBT treatment. Although the study is very preliminary, it highlights the utility of the sequenced approach to treating residual affective symptoms that has already been successfiiUy applied in recurrent unipolar depression.'*^
CBT for Relapse Prevention
One of the most important CBT studies in recent years was conducted by Lam and colleagues'*^ and involved 103 patients with bipolar I disorder who were randomized to receive treatment as usual or a 12- to 20-session individual cognitive therapy package. This package involved PE, cognitive therapy for depression, and identification of prodromes of relapse and action plans to avert relapse, as well as an emphasis on stabilizing routines. The investigators found that the group who received CBT had a significant reduction in relapse rates, fewer hospitalizations, fewer days in a bipolar episode, improved medication adherence, and improved psychosocial functioning. These positive outcomes were still seen even when the enhanced medication adherence was controlled for. A recent follow-up 2 years after treatment reported retention of gains."*^
Scott et al"** used a different 25-session CBT treatment package and randomized 42 bipolar disorder patients to CBT or a waiting list control condition. At 6-month follow-up, they observed decreased depressive relapses in the CBT group, as well as greater improvements in symptoms and functioning. In a more recent, much larger, multicentre RCT, Scott et al"*^ found a much more negative outcome associated with CBT. This study involved a more representative group of 253 bipolar disorder patients, of whom 40% had other psychiatric comorbidity, 47% had lifetime or current substance abuse, and 32% were in active episodes. Participants were randomized to receive usual care plus 20 CBT sessions (weekly for 15 weeks, then tapered over the next 11 weeks) or usual care alone. Usual care included medication, but notably, 16% of the sample were not receiving any mood stabilizer pharmacotherapy. Over the 18 months of follow-up, the researchers observed recurrence of bipolar disorder in 50% of usual care patients and in 53% of CBT patients. Mood severity ratings were almost identical in the 2 groups. Stratified analysis to control for patients having an acute episode at baseline showed similar results. Post hoc analysis revealed that bipolar disorder patients with fewer than 12 previous episodes did better with CBT than with usual care but that CBT was less effective in bipolar disorder patients with more episodes. This study suggests that, when other clinical factors (such as substance use and psychiatric comorbidity) are not controlled in an effort to make the patient sample more "ecologically valid," CBT may not be found as effective for patients with many previous episodes. In addition, including patients with active illness possibly weakened the impact of CBT since it is questionable whether CBT, rather than medication, would have been used as an initial treatment.
Zaretsky et al^" randomized 79 outpatients (52 with bipolar I disorder and 27 with bipolar II disorder) in full or partial remission to receive either 6 sessions of individual PE or 6 sessions of PE followed by 14 individual sessions of CBT. Although no differences in relapse rates were observed at 1-year follow-up from study entry, subjects receiving CBT in addition to PE experienced 50% fewer days in a depressed mood state and also had fewer antidepressant medication
Interpersonal and Social Rhythm Therapy
IPSRT was developed by Frank et al^' by integrating standard IPT for major depression with social rhythm therapy—a treatment approach that attempts to stabilize social and circadian rhythms based on the social zeitgeber hypothesis,^'^"'^ which posits that unstable or disrupted daily routines lead to circadian instability and affective episodes in vulnerable individuals. Although IPRST is a model-driven psychotherapy for bipolar disorder, the actual therapy itself contains many different elements, including PE, cognitive and behavioural interventions, self-monitoring of social rhythms, and IPT.
In a large, well-executed RCT, Frank et al compared 2 psychosocial interventions: IPSRT and an ICM for bipolar disorder.^'* Acutely ill individuals suffering from bipolar I disorder (« = 175) were randomized into 4 different treatment strategies: acute and maintenance IPSRT (IPSRT-IPSRT), acute IPSRT followed by maintenance ICM (IPSRT-ICM), acute ICM followed by maintenance IPSRT (ICM-IPSRT), or acute and maintenance ICM (ICM-ICM). The preventative maintenance treatment lasted 2 years. There were no differences observed among the treatment groups in time to stabilization or in the proportion achieving remission. In addition, overall recurrence rates were no lower in the IPSRT-IPSRT, group compared with the ICM-ICM group (in fact, the ICM-ICM group had a recurrence rate of 28%, which was 13% less than in the IPSRT-IPSRT group). However, participants assigned to acute phase IPSRT survived longer without an affective episode, regardless of the maintenance treatment, and had a higher regularity of social rhythms at the end of acute phase treatment. Ability to increase the regularity of social rhythms during acute treatment was associated with a reduced likelihood of recurrence during the maintenance phase. This study of IPSRT suggests that, although IPSRT does not dramatically reduce relapse rates alone, it can be used adjunctively to help slow the rate of relapse in a population with bipolar disorder. Further comparison of IPSRT to PE or CBT should help to define its relative efficacy and cost-effectiveness.
Psychotherapy for bipolar disorder is based on the premises that providing patients and their families with information about bipolar disorder will improve adherence to medication treatment, reduce substance abuse, enhance adaptive health beliefs (for example, sleep pattems), increase lifestyle regularity, and help patients to recognize early prodromal symptoms and thereby seek treatment sooner to avoid a fill relapse.
Psychosocial interventions focusing solely on adherence to medication treatment should not be underestimated, given the fact that 50% of individuals with bipolar disorder have at least one episode of medication noncompliance. Moreover, research has shown that acute lithium discontinuation increases the risk of rapid relapse and suicidal behaviour. As we have noted, PE is very effective for increasing treatment compliance in bipolar disorder. Therefore including PE in standard practice would be beneficial for most patients. However, PE does not affect all bipolar mood symptoms, and as such, the use of adjunctive FFT, CBT, or IPSRT is necessary to alleviate dysfunctional affective symptoms.
Specifically, CBT would be best used with patients who are euthymic since CBT is effective in reducing residual symptoms in bipolar disorder. It should be noted that, despite the mounting evidence that psychosocal interventions are beneficial to bipolar disorder patients, a limited number of RCTs validate these findings across different contexts. For example, studies involving comorbidity of eating disorders, anxiety, substance abuse, or attention-deficit hyperactivity disorder are very few. It is necessary to test these populations separately to gauge the issues that are illness-specific. In addition, treatment indications will likely differ according to the comorbid diagnosis, owing to the variance in symptoms across disorders.
Diagnostic issues in research are prevalent even when testing bipolar disorder alone. This is demonstrated by the fact that diagnoses of bipolar II disorder and rapid cycling are severely underrepresented in the research, as most studies assess patients with bipolar I disorder. As with comorbid diagnoses, it is possible that psychosocial interventions may affect the different subtypes of bipolar disorder differentially. Therefore, the bias toward testing bipolar I disorder patients needs to be rectified with more research on other bipolar types. Adolescent patients with bipolar disorder are also underrepresented in the psychosocial treatment literature. This could be for several reasons. First, general research in adolescent psychiatry can be problematic: a diagnosis will often change because psychiatric illnesses are not fully expressed during adolescence.^^ In this vein, it is difficult to recruit a true sample of adolescent patients without conducting longitudinal studies, which are not cost-effective. Second, adolescent patients are less reliable in terms of reporting symptoms and experience atypical symptoms." Despite these issues, it is still necessary to gain insight into how psychosocial treatments affect adolescents with bipolar disorder because earlier intervention could lead to reductions in relapse in adult years, which could result in an overall decrease in illness severity.
It is possible that psychotherapeutic interventions can target specific symptoms in bipolar disorder, such as insomnia and suicidality. CBT has demonstrated positive effects on treating primary insomnia and on suicide prevention. However, published data on this issue are relatively lacking. Additional studies are needed to tease out the effects of psychotherapy on these symptoms and to determine whether therapy has significant effects on any other primary symptoms of bipolar disorder. As with any intervention for psychiatric disorders, one needs to consider the feasibility of implementing specific treatments.
An issue with treating severely acute populations with FFT, CBT, or IPSRT is the practicality of delivering therapy that can last up to 9 months. This issue is likely more of a concern with patients suffering from severe mania than it is with patients suffering from depression because the pleasure experienced during mood elevation leads to an increased denial of the severity or existence of an illness and results in reduced acceptance of the need for treatment.*^" Studies have shown that the severity of mania and increased denial are significant predictors of nonadherence to medication treatment and, as such, may also translate to noncompliance with psychotherapeutic interventions.
Given the increased evidence that psychotherapy can improve treatment compliance and reduce symptoms in patients with bipolar disorder, beyond the effects of pharmacotherapy, the level of research is this area does not refiect the need. Future studies that include treatment matching based on diagnostic and biological variables, with cross-sectional and longitudinal designs, are needed to demonstrate the utility of psychosocial interventions in different contexts and over time.
The article above contains foundational information. Articles below contain optional updates.
- Zaretsky AE, Canadian Journal Of Psychiatry. Revue Canadienne De Psychiatrie, 2007 Jan; Vol. 52
Reflection Exercise #3
The preceding section contained information about the effectiveness of psychosocial interventions in bipolar disorder. Write three case study examples
regarding how you might use the content of this section in your practice.
What is the social zeitgeber hypothesis? Record the letter of the correct answer the