Bipolar Disorder Often Mistaken for Depression in Primary Care
News Author: Megan Brooks
CME Author: Désirée Lie, MD, MSED
March 8, 2011
As many as 20% of adults being treated for depression in primary care may have undiagnosed bipolar disorder (BD), suggests a new study from the United Kingdom published online February 3 in the British Journal of Psychiatry.
It's likely, say the researchers, that many of these patients are receiving antidepressant monotherapy. "For people who in fact have undiagnosed bipolar disorder, these drugs may be at best unhelpful and at worst harmful," lead researcher Daniel J. Smith, MD, MRCPsych, of the Department of Psychological Medicine and Neurology, Cardiff University School of Medicine and University Hospital of Wales, cautions in a statement.
BD can be difficult to diagnose and is often misdiagnosed as recurrent major depressive disorder, Dr. Smith and colleagues point out in their report. They invited 3117 adults from South Wales who were being treated for depression by their primary care physician to participate in their study.
A total of 576 (18.5%) agreed to participate and completed 2 screening instruments for BD (the Hypomania Checklist [HCL-32] and the Bipolar Spectrum Diagnostic Scale [BSDS]). A total of 370 of them were invited to take part in a one-on-one comprehensive diagnostic and clinical assessment, 154 agreed, and 29 (18.8%) met Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) (DSM-IV) criteria for BD.
The researchers generated 3 estimates of the percentage of patients in primary care with a diagnosis of unipolar depression who may, in fact, have unrecognized BD.
Their most conservative estimate of 3.3% is based on the strictest assumption that all individuals who dropped out of the study by not completing the questionnaire or being assessed did not have BD.
"This seems unlikely and therefore provides an extreme lower bound for the proportion of people with undiagnosed bipolar disorder in this population," the researchers write.
Their least conservative estimate of 21.6% assumes that all individuals who did not drop out would have responded in the same way as those who were assessed. The researchers' more conservative midestimate of undiagnosed BP in depressed primary care patients is 9.6%.
Therefore, between 3.3% and 21.6% of primary care patients with depression may have unrecognized BD, Dr. Smith and colleagues note. "It's therefore important," says Dr. Smith, "that the possibility of undiagnosed bipolar disorder is given greater recognition in primary care, and that GPs [general practitioners] are supported in developing strategies to ensure that their patients with depression receive the correct diagnosis."
The findings also suggest that the HCL-32 and BSDS screening questionnaires when used in primary care settings may have "limited utility" in terms of detecting DSM-IV–defined BD. The positive predictive values were 50.0% and 32.1%, respectively. However, these instruments may be useful for identifying broader definitions of BD, the researchers say.
In line with other studies, they also found that subthreshold features of BD were relatively common in the cohort of depressed primary care patients and were associated with a more morbid course of illness and greater psychosocial and quality-of-life impairments.
"These findings have important implications for the classification, assessment, and treatment of large numbers of people with depression managed in both primary and secondary care settings," they conclude.
The study was funded by an MRC/Welsh Assembly Government Partnership Award. Dr. Smith is funded by a Postdoctoral Fellowship from the National Institute of Health Research. Dr. Smith discloses having received honoraria for speaking at educational meetings organized by AstraZeneca and Lilly.
Br J Psychiatry. Published online February 3, 2011.
The Mayo Clinic’s Web page on bipolar disorder provides in-depth information for patients about the disease, its diagnosis, and management and provides links to other resources for patients.
BD is often under-recognized in clinical practice and is often misdiagnosed as recurrent major depressive disorder. The mean delay for the diagnosis of BD is estimated as 10 years from onset of the first symptoms. There may be a high prevalence of undiagnosed BD among patients with unipolar depression, and the HCL-32 and BSDS may be useful for screening for manic symptoms in a primary care population.
This 2-phase screening study was conducted in primary care to determine the prevalence of BD among patients with unipolar depression and the usefulness of the 2 screening instruments to diagnose manic symptoms.
• Included were patients with a current diagnosis of unipolar depression from 11 of 45 primary care practices in a single health board in South Wales. These patients had detailed diagnostic, clinical, psychosocial, and quality-of-life data.
• Participants were invited to complete the HCL-32 and BSDS questionnaires.
• 576 participants completed the questionnaires and were divided into 2 groups: high scorers (≥ 14 on the HCL-32 or ≥ 13 on the BSDS; n = 411) and low scorers (< 14 on the HCL-32 or < 13 on the BSDS; n = 165).
• All of the low scorers and half of the high scorers participated in a comprehensive diagnostic and clinical assessment.
• The assessments included the Mini International Neuropsychiatric Interview and a structured assessment of psychosocial and medical history and medications.
• A 2-phase sampling technique was used to estimate the prevalence of BD among the patients.
• The investigators assessed the usefulness of the HCL-32 and BSDS with receiver operating characteristic curves.
• 116 of 154 participants interviewed had a history of DSM-IV–defined major depressive disorder, and 29 (18.8%) satisfied DSM-IV criteria for BD.
• With the 2-phase sampling technique, the estimate of the prevalence of BD among those with unipolar depression ranged from 3.3% to 21.6%, with a midpoint estimate of 9.6%.
• Both the HCL-32 and the BSDS could distinguish between BD and major depressive disorder with reasonable precision (area under the curve was 0.81 for the HCL-32 and 0.71 for the BSDS).
• The optimal cutoff values were a score of 18 or higher on the HCL-32 and a score of 12 or higher for the BSDS, but the positive predictive value was low (50% for the HCL-32 and 32.1% for the BSDS).
• In the sample of 116 participants, 60 had major depressive disorder with manic symptoms, and 56 had major depressive disorder only.
• Patients with major depressive disorder accompanied by manic symptoms were younger, had a younger age of onset of depression, reported more frequent symptoms of depression, and were more likely to have comorbid alcohol dependence.
• They also had greater difficulty with functioning and relationships and had lower quality-of-life scores.
• Overall, patients with major depressive disorder and manic symptoms had more severe depression, poorer psychosocial functioning, and worse quality of life vs patients with major depressive disorder only or those with no subthreshold manic symptoms.
• The authors concluded that the prevalence of BD among patients with unipolar depression ranged from 3.3% to 21.6%, and up to 1 in 30 patients in primary care with unipolar depression may have unrecognized BD.
• They also noted that both the HCL-32 and SDS had a low positive predictive value for diagnosing DSM-IV–defined BD.
• The prevalence of BD among patients with unipolar depression ranges from 3.3% to 21.6%.
• The HCL-32 and the BSDS have low positive predictive values for diagnosing BD among patients with unipolar depression.