Bipolar Disorder Prevalence Hidden by Diagnostic Thresholds

BETHESDA, Md., May 8 -- There may be twice as many patients with bipolar spectrum disorders as generally suspected, according to a national survey.

The survey found that 4.4% of Americans have had a form of bipolar disorder at some point in their lifetime, whereas the prevalence of bipolar I and II combined over a lifetime is typically cited at about 1.9%.

The difference was attributed to the 2.4% prevalence of "subthreshold" bipolar disorder, Kathleen R. Merikangas, Ph.D., of the National Institute of Mental Health here, and colleagues, reported in the May issue of the Archives of General Psychiatry.

"Subthreshold bipolar disorder is common, clinically significant, and underdetected in treatment settings," they wrote. "Explicit criteria are needed to define subthreshold bipolar disorder for future clinical and research purposes."

The findings explain the "large discrepancy" between rates in large-scale community surveys and in prospective longitudinal studies, they said.

Previous surveys have had criteria that were too restrictive for symptoms and diagnosis thresholds to pick up bipolar disorder in the general population, Dr. Merikangas and colleagues said, "particularly in young adults, when the disorder is in evolution."

So in the National Comorbidity Survey Replication study, they included broader criteria along with the traditional Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) definitions of mania, hypomania and major depression.

The criterion for bipolar I disorder was ever having had a manic episode. For bipolar II disorder, the criteria were whether patients ever having had a hypomanic but not manic episode and ever having had an episode of major depression.

The criteria for subthreshold bipolar disorder were:

• Recurrent subthreshold hypomania (at least two criterion B symptoms and all other criteria for hypomania) in the presence of major depression.
• At least two episodes of hypomania in the absence of recurrent major depression.
• Recurrent subthreshold hypomania in the absence of major depression.

These definitions were applied to responses from the nationally representative sample of 9,282 adults interviewed using the structured WHO Composite International Diagnostic Interview.

Those who met criteria for any core psychiatric disorder and a random sample of the other participants answered additional questions. Bipolar diagnosis cases with plausible organic causes were excluded.

Among the prevalence findings, the researchers reported:

• 1.0% had bipolar I disorder during their lifetimes.
• 1.1% had bipolar II disorder during their lifetimes.
• 2.4% had subthreshold bipolar disorder during their lifetimes.
• 0.6% had bipolar I disorder during the prior year.
• 0.8% had bipolar II disorder during the prior year.
• 1.4% had subthreshold bipolar disorder in the prior year.

Other results showed that the spectrum concept of bipolarity is valid, they said.

"The direct association between increasingly restrictive definitions of bipolar disorder and the indicators of clinical validity, including number of episodes, chronicity, symptom severity, impairment, comorbidity, and treatment, provides evidence of the underlying dimensional nature of bipolar illness," they wrote.

Age at onset was higher on average with decreasing disease severity (18.2 for bipolar I disorder, 20.3 for bipolar II disorder, and 22.2 for subthreshold bipolar disorder).

Most bipolar patients had other comorbid psychiatric disorders, but the proportion was lower among those with subthreshold bipolar disorder than threshold bipolar I and II disorders (88.4% versus 95.8% to 97.7%).

Severity of work impairment rose from 19.8% for subthreshold bipolar disorder to 47.5% for bipolar II to 62.3% for bipolar I disorder.

Severe manic or hypomanic episodes were more common with rising bipolar severity (31.5% for subthreshold bipolar, 55.4% for bipolar II, and 70.2% for bipolar I disorder). Likewise, estimated average lifetime episodes were 77.6 for bipolar I, 63.6 for bipolar II, and 32.0 for subthreshold bipolar disorder.

Treatment for emotional problems at any point in their lifetime was common, and more so for bipolar I and II (89.2% to 95.0%) than for subthreshold bipolar disorder (69.3%).

However, only a few received appropriate medication (25.0% for bipolar I, 15.4% for bipolar II, and 8.1% for subthreshold bipolar disorder). Appropriate maintenance medication for currently asymptomatic patients was even lower (17.9%, 15.6%, and 3.2%, respectively).

Inappropriate medication use -- primarily antidepressants in the absence of antimania agents -- was considerably more common (31.4% for 12-month cases and 25.1% for asymptomatic lifetime cases).

"The high use of inappropriate medications is a concern given the dangers associated with the use of antidepressants in the absence of mood stabilizers to treat bipolar disorder," Dr. Merikangas and colleagues wrote.

The study likely underestimated the prevalence of subthreshold bipolar disorder, they noted. The structured interview precluded using lower thresholds for mania and depression diagnostic criteria and gathering information on mixed episodes, rapid cycling, and brief episodes.

Nevertheless, "the present results reinforce the argument of other researchers that clinically significant subthreshold bipolar disorder is at least as common as threshold bipolar disorder," they wrote.

The results also "strongly argue for inclusion of the bipolar disorder spectrum concept in the diagnostic classification system," the investigators added.

"More comprehensive screening of bipolar symptoms is needed in people seen for treatment of other Axis I disorders," they concluded.

One of the researchers reported being a consultant to or on the advisory board of Abbott Laboratories, AstraZeneca, Bristol-Meyers Squibb, Forest Laboratories, GlaxoSmithKline, Janssen Pharmaceutica, Eli Lilly, Novartis, Organon, Pfizer, Shire, UCB Pharma, and Wyeth-Ayerest.