Bipolar II Disorder: Understudied and Underdiagnosed

Alternating states of mania and melancholia are among the earliest described human diseases, first noted by ancient Greek physicians, philosophers, and poets (14). Hippocrates (460–337 B.C.E.), who formulated the first known classification of mental disorders, systematically described bipolar mood states: melancholia, mania, and paranoia. More than 2 millennia later, Emil Kraepelin, recognized as one of the founders of modern psychiatry, described manic-depressive illness in 1899 as a singular disease characterized by alternating cycles of mania and melancholia. Kraepelinian nosology, however, focused on episodic recurrences rather than the polarity of episodes. Thus, his concept subsumed several contemporary diagnostic categories, including bipolar I disorder (recurrent mania and depression), bipolar II disorder (recurrent hypomania and depression), and recurrent major depressive disorder (recurrent depression). In 1957, Karl Leonhard proposed dividing Kraepelin’s manic-depression into bipolar disorder and unipolar disorder, narrowing the definition of bipolarity to comprise illness with both manic and depressive symptoms (15). These historic formulations provide background for our modern concepts of mood disorders in general and bipolar disorders more specifically.

Bipolar II disorder was first proposed by David Dunner in the 1960s (16, 17). When examining a cohort of individuals with mood disorders, he identified a subgroup of patients with recurrent episodes of depression who also had histories of at least one period of hypomania and a strong family history of bipolar disorder. This subgroup was found to have a different course of illness compared with those with recurrent depression and a history of mania (bipolar I disorder). Thanks to this work and follow-up research, bipolar II disorder was recognized as a distinct disorder, separate from bipolar I disorder. It finally entered the DSM lexicon in 1994 in DSM-IV (18) and the ICD-10 even more recently.

Some groups raise concerns about the definitional integrity of bipolar subtypes and validity of bipolar II disorder as a separate disorder, noting poorly defined syndromic boundaries, limited biologic and treatment factors separating bipolar II disorder from bipolar I disorder, and overlapping genetic risk (19). However, a preponderance of evidence establishes bipolar II disorder as a distinct entity. Descriptive studies identify clinical and demographic features specific to bipolar II disorder (20, 21). Family studies show distinct heritability patterns in those with the bipolar II disorder phenotype (16, 22), findings that are supported by large population-based familial aggregation studies showing differences in patterns of bipolar I disorder and bipolar II disorder heritability (23). Differential treatment responsivity to agents such as antidepressants (24–26) underscores the validity of considering bipolar I and II disorders as distinct conditions.

In DSM-5 field trials to assess reliability of diagnoses, bipolar I disorder was among the most recognizable (κ=0.73), but bipolar II disorder fell in the acceptable range (κ=0.40) and well above major depressive disorder (κ=0.28), supporting the reliability of this diagnostic entity (27). Bipolar II disorder diagnosis also reliably separates from bipolar I disorder in careful clinical interviews (28).

Bipolar II disorder is operationalized as at least one lifetime hypomanic episode and one major depressive episode (6). Beginning with the fifth edition of DSM, a diagnosis of mania or hypomania requires changes in both mood and energy, reflecting a growing understanding that these illnesses are likely manifestations of dysregulated energy expenditure as much as mood (29). As a reminder, hypomania requires a minimum symptom duration of 4 days, whereas mania requires a minimum duration of 1 week. There is no maximum duration specified for either mania or hypomania. Definitionally, hypomania cannot cause impairment, whereas mania must.

In psychiatry, all diagnoses are a one-way road. If someone has ever met criteria for a manic episode, they will continue to carry the diagnosis of bipolar I disorder—even without further manic episodes. Similarly, if someone has a distant episode of hypomania and at least one prior major depressive episode (but no manic episodes), they would be considered to have bipolar II disorder, even in the absence of subsequent hypomanic episodes meeting symptom and duration criteria. Notably, only about 5% of adults with bipolar II disorder subsequently develop bipolar I disorder, further arguing for the stability of this bipolar disorder subtype over time (30). Table 1 summarizes how episode types are combined to yield mood disorder diagnoses.

Despite the specificity of bipolar II disorder diagnostic criteria, clinicians struggle to identify it accurately in practice. Bipolar II disorder is often either missed or misdiagnosed, resulting in a delay in diagnosis of more than 10 years (2, 31). Difficulties in accurate diagnosis arise from several sources. First, DSM-5 criteria for the depressive phase of bipolar II disorder are identical to those required for a major depressive episode in the context of major depressive disorder, which make bipolar II disorder and major depressive disorder cross-sectionally indistinguishable when a patient is depressed (see Table 1). This is particularly notable, as diagnoses of major depressive disorder make up a substantial percentage of incorrect diagnoses (32). Second, hypomania (which, by definition, is a less severe form of mania) may be difficult for patients to distinguish from a “normal” mood state accompanied by extra energy and good mood. Whereas individuals with bipolar I disorder can easily recognize and describe prior manias (which are, by definition, more dramatic and memorable), many individuals with bipolar II disorder struggle to identify episodes of hypomania retrospectively, especially during a depressive episode. Third, mixed hypomanic mood states are very common in bipolar II disorder (21) and, in fact, more common than euphoric hypomanic states (33). Mixed mood states are characterized by the presence of symptoms of opposite polarity during a depressive or hypomanic episode. In an episode of mixed hypomania, patients might believe that they are simply irritable and angry in the context of depression rather than recognize the additional hypomanic symptoms that warrant a diagnosis of mixed hypomanic state. Mixed-mood episodes can contribute to diagnostic confusion, especially between bipolar disorder and unipolar depression, as depressive episodes with mixed features can occur in both disorders. Fourth, the degree of impairment is a dividing line between hypomania and mania (no impairment for hypomania, definite impairment for mania), a construct that can be difficult to operationalize and measure, especially in clinical practice. Finally, the primary reason why patients with bipolar II disorder seek care is depression. Patients rarely present for treatment in the midst of a hypomanic episode, a mood state that is either perceived as ego-syntonic or simply not identified as part of their illness (34). As a result, many individuals with bipolar II disorder fail to recall, recognize, or report histories of hypomania, contributing to a misdiagnosis of major depressive disorder.

Accurate diagnosis relies on the careful eliciting of patient history, especially for the sometimes elusive diagnosis of bipolar II disorder. To improve diagnostic accuracy, clinicians should systematically screen all patients with depression for bipolar disorder, ideally using an evidence-based screening tool. Several well-validated screening tools for bipolar disorder are available, including the Mood Disorder Questionnaire (35), the Hypomania Checklist (36), and the Bipolar Spectrum Diagnostics Scale (37). All extant screening tools are imperfect, having sensitivities and specificities less than 100% (38). No tool screens specifically for bipolar II disorder. Screening tools should not be used to make a clinical diagnosis. Despite these caveats, screening tools can be very useful, when part of a comprehensive approach to care, to identify individuals who may have bipolar disorder.

The DSM focuses on categorial diagnoses (i.e., thresholds for the absence or presence of disease). In parallel to this framework, many have argued to conceptualize bipolar disorders along a continuous spectrum of illness. Thus, the term bipolar spectrum is used to describe both the spectrum of severity across symptoms of bipolar disorder and the combinations of mood symptoms with manic-hypomanic and depressive components (39). Some refer to bipolar II disorder as a part of the bipolar spectrum. These concepts reflect a growing awareness that dimensional descriptions of mood disorders may better map onto continuous biological markers of disease compared with DSM’s categorical approach, but with ongoing debates about diagnostic boundaries (19, 40, 41). Notably, conceptualizations of bipolar disorder as a spectrum condition versus discrete diagnostic categories (e.g., bipolar I disorder or bipolar II disorder) are not mutually exclusive but rather speak to ongoing efforts to understand and optimally describe the phenomenology of bipolar disorder.

Bipolar II disorder is a relatively common disorder affecting approximately 0.4%–1% of the population (42, 43). Depression dominates the course of bipolar II disorder (44), a clinical feature that is common to both bipolar I and bipolar II disorders (45). Depression is the primary driver of morbidity in bipolar II disorder (43) and is the phase of the disorder that is most troubling to patients. Most treatments for bipolar II disorder primarily target depression (46).

Before the era of efficacious treatments, Kraepelin noted that the course of illness for patients with bipolar disorder generally progresses into more persistent and severe depression with aging (14). Although he was primarily referring to manic-depressive illness, which we call bipolar I disorder today, the same principle applies to patients with bipolar II disorder. In the National Institute of Mental Health Collaborative Study, which included long-term follow-up of up to 20 years, patients with bipolar II disorder experienced fewer well intervals and more depressive episodes over time (11, 44). Similar patterns have also been noted in other studies (47).

Clinical features predictive of a bipolar disorder diagnosis (vs. major depressive disorder) include earlier age at onset (age younger than 25 years), more recurrent episodes (more than five episodes), the presence of subthreshold hypomanic or mixed symptoms, a family history of bipolar disorder, atypical depressive symptoms, comorbid substance use disorder, and nonresponse to antidepressant medications (48–50). Although none of these features are pathognomonic of bipolar disorder, their presence should raise suspicion of a bipolar disorder diagnosis. These characteristics do not distinguish bipolar I disorder from bipolar II disorder.

There is a long-standing debate in the literature about whether patients with bipolar II disorder experience the same degree of disability as those with bipolar I disorder. Bipolar II disorder was previously—and incorrectly—labeled a “less severe” version of bipolar I disorder. In fact, studies consistently show comparable disease burden in bipolar I and bipolar II disorders (10, 20, 21). A recent Swedish study comparing almost 9,000 patients with bipolar I disorder (N=4,806) and bipolar II disorder (N=3,960) found that, compared with those with bipolar I disorder, individuals with bipolar II disorder reported higher rates of depressive episodes, illness onset at a younger age, and significantly higher rates of psychiatric comorbidity (anxiety disorders, eating disorders, and attention-deficit hyperactivity disorder [ADHD]). In this Swedish sample, no differences were noted in substance abuse between patients with bipolar I disorder and those with bipolar II disorder (20).

High rates of psychiatric and medical comorbid conditions in bipolar II disorder further compound the challenges of differential diagnosis and treatment (51). In the largest study to date assessing comorbid psychiatric conditions in bipolar II disorder, 83% of individuals had at least one psychiatric condition in addition to bipolar II disorder. Over 50% had had three or more comorbid diagnoses (42). Rates of co-occurring anxiety disorders are especially high in individuals with bipolar II disorder, with up to 89% experiencing at least one anxiety disorder (52). Some investigators have argued that anxiety itself may be part of a bipolar prodrome (53), placing youths with early onset anxiety symptoms at risk for developing bipolar spectrum disorders. Impulse control disorders, including ADHD and intermittent explosive disorder, affect approximately half of individuals with bipolar II disorder (54, 55). Alcohol use disorder affects a third of those with bipolar II disorder (56).

Because of an overlap in phenomenology (e.g., impulsivity, affective lability, dysphoria, and suicidal behaviors), distinguishing between bipolar II disorder and borderline personality disorder may be especially challenging (57, 58). To distinguish bipolar II disorder from borderline personality disorder, it is important to systematically assess for nonoverlapping diagnostic criteria; for instance, identity disturbance, fears of abandonment, and chronic feelings of emptiness that are specific to borderline personality disorder (6). Also, mood changes in bipolar II disorder are of a longer duration than the brief emotional fluctuations characteristic of borderline personality disorder (“climate vs. weather”), but, in practice, it may be difficult to sort out. Notably, these disorders are not mutually exclusive, with up to 20% of individuals with bipolar II disorder also meeting criteria for borderline personality disorder (59). Conversely, up to 40% of those with borderline personality disorder are misdiagnosed as having bipolar I disorder or bipolar II disorder (59, 60). Thus, distinguishing bipolar II disorder from borderline personality disorder requires a careful clinical assessment of both past and current symptomatology, including careful consideration of symptom clusters specific to each disorder.

Medical comorbidity is the rule rather than the exception with bipolar II disorder. More than 90% of individuals with bipolar II disorder have significant medical comorbidity (61, 62). In one report, 22% of individuals with bipolar I disorder and bipolar II disorder had metabolic disorders, 19% had cardiovascular disease, 19% had thyroid disorders, and 8% had neurological diseases (63). Although rates of medical comorbidity in bipolar I disorder and bipolar II disorder appear to be comparable, rates of migraine headaches (64) and irritable bowel syndrome (65) appear to be higher in individuals with bipolar II disorder. It is interesting that proinflammatory medical conditions are especially common in patients with bipolar II disorder (66), raising the possibility of shared pathophysiology related to immunological dysfunction (67).

Suicide is a significant risk for all patients with bipolar disorder, and historically, patients with bipolar I disorder were viewed as a higher risk than bipolar II disorder because of the extremities of mania. However, data from a number of sources support that suicide risk is high across all patients with bipolar disorder, with relatively little difference in risk between patients with bipolar I disorder and those with bipolar II disorder (12). Older studies suggested that this risk may be higher for patients with bipolar II disorder than for those with bipolar I disorder (68); indeed, a recent Swedish bipolar registry database noted that the rate of suicide attempts were significantly higher in patients with bipolar II disorder compared with patients with bipolar I disorder, although no data on completed suicides were provided (20). Overall, reports from the International Society for Bipolar Disorders Task Force on Suicide found that the risk for suicide was estimated at 164 out of 100,000 patients per year among those with bipolar disorder versus 10 out of 100,000 individuals per year among the general population (69, 70).

Treatment guidelines for bipolar disorder often only give a passing nod to distinguishing appropriate treatments for bipolar I disorder versus bipolar II disorder. One recently published set of consensus recommendations was unusual in making a point to distinguish the evidence base for bipolar I disorder versus bipolar II disorder (46). A collaborative effort of Canadian experts and the International Society for Bipolar Disorders, these guidelines have a separate section devoted to bipolar II disorder that represents expert opinions rather than evidence-based recommendations: They clearly state that one cannot apply the findings of studies on patients with bipolar I disorder to the management of patients with bipolar II disorder and conclude that there are too few controlled studies of patients with bipolar II disorder to make detailed evidence-based recommendations or develop evidence-based treatment algorithms. The relative paucity of data makes treatment planning more challenging; however, an online patient decision support tool may help individuals with bipolar II disorder navigate this complex decision-making process (71). Below is a brief overview of our current understanding of recommended treatment for patients with bipolar II disorder. Table 2 summarizes treatments that have at least some evidence supporting their use for bipolar II disorder.

Bipolar II disorder is a relatively common disorder, affecting one out of every 100 individuals. Its prevalence, morbidity, and mortality are comparable with those of bipolar I disorder. Evidence supports conceptualizing bipolar II disorder as a distinct phenotype, separable from both bipolar I disorder and major depressive disorder. However, despite being reliably diagnosed in DSM-5 field trials, bipolar II disorder is frequently misdiagnosed in practice, resulting in a decade-long lag between onset of symptoms and appropriate diagnosis.

Compared with bipolar I disorder and major depressive disorder, far less is known about how to treat bipolar II disorder. Expert consensus guidelines (46) recommend quetiapine, lithium, and lamotrigine as first-line maintenance treatments for bipolar II disorder and quetiapine as first-line treatment for acute depression. Limited data are also available supporting antidepressant monotherapy and lumateperone as treatments for acute bipolar II disorder depression. In summary, bipolar II disorder is a neglected condition, causing unnecessary suffering in those who are misdiagnosed or for whom appropriate treatments are unclear. More research is urgently needed to improve identification and treatments for bipolar II disorder.