Although the diagnosis of bipolar disorder is defined by the occurrence of mania or hypomania, in fact, depressive episodes and symptoms tend to dominate the course of the illness (1). However, treatments for bipolar depression are much less well defined than those for mania. To highlight this point, contrast the number of treatments approved by the U.S. Food and Drug Administration for mania (N=9; lithium, chlorpromazine, divalproex, olanzapine, risperidone, quetiapine, aripiprazole, ziprasidone, and carbamazepine) with that for bipolar depression (N=2; quetiapine and olanzapine-fluoxetine combination). Consequently, treatment of bipolar depression typically relies on off-label medication use, particularly of antidepressants, or nonpharmacological interventions. The efficacy of these interventions and how they affect the subsequent course of bipolar illness remain areas of limited information and moderate controversy. Two articles from the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) published in this month’s Journal address these shortcomings in our knowledge base.

STEP-BD is a large National Institute of Mental Health (NIMH) clinical research program designed to study treatment effectiveness with both naturalistic and randomly assigned treatment protocols (2). Fifteen sites from across the United States participated in this program. One goal of STEP-BD was to acquire treatment, course, and phenomenological information that may be more clinically relevant to everyday practitioners than is often obtained from large, industry-sponsored studies that, by requirement, exclude populations of patients (e.g., those with comorbid conditions) that commonly occur in general clinical practice. Like all clinical research efforts, the STEP-BD sample was not perfectly generalizable; for example, minority patients are underrepresented, and subjects presenting to academic centers are overrepresented. Nonetheless, despite these limitations, results from STEP-BD will provide reasonable facsimiles of many psychiatrists’ practices such that the results reported will be clinically applicable.

A major strength of STEP-BD is that it provides an unusually large sample of well-characterized bipolar patients, which permits evaluation of treatment response and course of illness in important understudied subgroups of bipolar patients. One of these subgroups is patients in the depressive phase of illness with concurrent manic symptoms that do not meet criteria for a mixed state. Defining the boundaries of mixed states has been a controversial topic for many years, although most discussion has focused on dysphoric mania (i.e., manic episodes with concurrent depressive symptoms) rather than depression with subsyndromal mania (3). This subgroup of patients is clinically relevant in that, at least hypothetically, one might presume that they would be at a higher risk for developing full (syndromal) mania in response to antidepressants compared with depressed patients with no manic symptoms. With these considerations in mind, as reported in this issue, Goldberg and colleagues drew 335 patients from the first 2,000 subjects recruited into STEP-BD in order to examine whether antidepressants added to mood stabilizer treatment hastened recovery in bipolar depression complicated by manic symptoms. Additionally, the investigators examined whether this subgroup of patients was at particular risk of developing antidepressant-induced mania. Patients were treated naturalistically within the STEP-BD network, although with clear treatment algorithms to guide participating clinicians. All patients received mood stabilizers or atypical antipsychotics, and then those who also received antidepressants were contrasted with those who did not. The study found that recovery from depression was independent of whether or not patients received adjunctive antidepressant treatment. These results mirror another recent publication from STEP-BD that found no advantage of adding antidepressants to mood stabilizers in the treatment of bipolar depression without concurrent manic symptoms (4). These findings are also consistent with a double-blind, placebo controlled study of bipolar depression that found that if lithium was dosed to a serum level of at least 0.8 meq/liter, then the addition of an antidepressant (paroxetine, imipramine) provided no additional benefit in symptom improvement (5). In contrast to the lack of association between antidepressant use and clinical recovery, the article observed a statistically significant association between antidepressant use, the number of baseline manic symptoms, and the endpoint mania rating. However, this association was not linear (i.e., more manic symptoms at baseline did not necessarily predict greater mania ratings with antidepressant use at endpoint), and the specific clinical meaning of the statistical association appeared to be quite complex. Unfortunately, the questions of which bipolar depressed patients are at particular risk for antidepressant-induced mania and how often this occurs in any bipolar sample remain unanswered, even in this relatively large data set. Nonetheless, this article adds to the growing evidence that in the setting of adequate therapeutic mood stabilizer prescription, the addition of antidepressants appears to provide little additional benefit for depressed bipolar patients but may impart a risk of switching to mania.

Given the apparent relative lack of efficacy of antidepressants in the treatment of depressed bipolar patients, other alternatives are clearly needed. Psychotherapies provide treatment options that presumably will not induce mania. Previously, using the STEP-BD sample, Miklowitz and colleagues demonstrated that the addition of intensive psychotherapy (cognitive behavior therapy, interpersonal and social rhythm therapy, and family-focused therapy) to standard mood stabilizer treatment was associated with better symptomatic recovery than mood stabilizer treatment without these treatments (6). In this month’s issue, Miklowitz and colleagues extend their prior work by examining the effects of intensive psychotherapies on functional recovery from bipolar depression. The authors compared these three intensive psychotherapies (cognitive behavior therapy, interpersonal and social rhythm therapy, and family-focused therapy) to "collaborative care"; the latter consisted of three sessions focusing on helping patients implement self-management tools (e.g., mood monitoring). Four domains of function were examined, including relationships, work/role performance, recreational activities, and life satisfaction during a 9-month period. All patients were receiving standard pharmacotherapies. The patients who received intensive therapy demonstrated better total functioning, relationship functioning, and life satisfaction compared to those receiving collaborative care. The three types of intensive psychotherapies did not appear to provide significantly different benefits. Additionally, no differences between groups were observed for work/role functioning and recreational activities. A previous study in first-episode patients suggested that different components of recovery would require different interventions (7), and the current study by Miklowitz et al. supports this assertion, namely, that intensive psychotherapies may lead to improvements in some aspects of functional recovery that medications do not alone benefit. Although the impact of intensive psychotherapies on functional improvement demonstrated modest effect sizes compared with collaborative care, given the significant functional impairment associated with bipolar disorder, even modest gains are clinically meaningful.

These two articles from STEP-BD highlight the primary strengths of this large NIMH-sponsored program, namely, the ability to look at important bipolar subgroups and contrast treatments in "real-life" samples of bipolar patients. These specific results inform and challenge investigators and clinicians to continue to advance clinical care of people suffering with bipolar disorder. After decades of languishing with minimal research and few treatment options, the investment of the federal government in projects like STEP-BD and large psychopharmacology companies’ interest in developing new treatments for bipolar disorder that has occurred in the last decade suggest that the future for the treatment of bipolar disorder will be steadily becoming brighter.