Bipolar disorder may be underdiagnosed in epidemiological surveys, say researchers who studied data from the World Mental Health (WMH) surveys.
About a quarter of patients potentially meeting the criteria for bipolar disorder may have been inappropriately screened out, say Elie Karam (Institute for Development, Research, Advocacy & Applied Care, Beirut, Lebanon) and co-workers.
They attribute this to the wording of five key questions in the Composite International Diagnostic Interview (CIDI), which is used in the WMH surveys. These questions, they believe, are worded in a way that may produce false–negative responses, and because these particular questions are “gate” questions, a negative answer will cause the interviewer to skip the rest of the section.
The researchers stress that the participants they identified as being inappropriately screened out had given a positive response to “stem” questions concerning extended bouts of euphoria or irritability.
Their study covered eight WMH surveys carried out in five geographical areas and involving 47,552 people, 71.4% of whom actually completed an interview. Of these, 69.5% responded negatively to all stem questions (reference group), 6.8% responded positively to a stem question regarding euphoria but were then screened out and 18.9% responded positively to an irritability stem question but were screened out. The other 3.6% of participants had a positive screen for bipolar disorder.
The characteristics of the patients who may have been inappropriately screened out were intermediate between that of the reference group and those with a positive screen. For example, the rates of major depression were 29.1% in the euphoric subgroup and 26.0% in the irritable subgroup, compared with 6.4% in the reference group and 48.6% to 51.3% in participants with a positive screen.
The same pattern emerged for age at onset, number of episodes, suicide attempts and comorbid anxiety, disruptive behaviour and substance use disorders.
Karam et al note that the problem of potential underdiagnosis is not restricted to the CIDI, with problems being reported for other tools. Most of the difficulties “reflect a tendency to equate bipolarity with its more disruptive type (Bipolar I)”, they say, with bipolar II disorder more likely to be overlooked.
“All experienced clinicians would agree that the diagnosis of bipolar disorder, especially hypomania, is not always easy to elicit and might need careful and repeated interviewing as well as calling on other informants (spouse, relatives),” they write in the Journal of Affective Disorders.
The team concludes that review of screening tools “will undoubtedly prove correct what many clinical studies have been showing lately, namely that bipolar disorder is more common than results from what most epidemiologic studies have shown so far”.
This article first appeared on ‘Med Wire News’ on 4 February 2014.