Bipolarity in depressive patients without histories of diagnosis of bipolar disorder and the use of the Mood Disorder Questionnaire for detecting bipolarity

Compr Psychiatry. 2008 Sep-Oct;49(5):469-75. doi: 10.1016/j.comppsych.2008.01.002. Epub 2008 Jun 3.


Objectives: The present study was performed to evaluate the frequency of bipolar disorders among patients (a) presenting with depressive episodes but (b) who have never been diagnosed with bipolar disorder (c) in routine clinical practice in Korean subjects and to identify which clinical features were helpful in discriminating bipolar patients from unipolar patients. In addition, authors assessed the practical use of the Mood Disorder Questionnaire (MDQ) to distinguish bipolar from unipolar disorder in these subjects and tested whether modifications of the MDQ scoring could improve its performance.

Methods: We evaluated consecutive patients who satisfied the inclusion criteria of a current depressive episode, plus at least one previous depressive episode. Subjects were interviewed for diagnosis using the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders IV after completing the MDQ. To improve assessment of hypomania history, the interviewer made strenuous efforts to explore a possible history of hypomania, and patient-derived data were supplemented by information from family members or close relatives.

Results: Fifty-nine patients (53.2%) were classified as having bipolar disorder, leaving a group of 52 (46.8%) with unipolar depression. Among bipolar disorders, 1.8% (n = 2) had bipolar I disorder; 29.7% (n = 33), bipolar II disorder; 6.3% (n = 7), bipolar III disorder (history of antidepressant-induced hypomania without spontaneous hypomanic episode); and 15.3% (n = 17), bipolar disorder not otherwise specified (1-3 days brief hypomania). Postpartum depression (relative risk [RR] [95% confidence interval {CI}], 2.00 [1.23-3.24]), early age of onset (RR [95% CI], 1.85 [1.30-2.64]), mood lability (RR [95% CI], 1.85 [1.30-2.64]), brief depressive episode (RR [95% CI], 1.66 (1.16-2.37]), bipolar family history (RR [95% CI], 1.62 [1.08-2.43]), history of suicide attempt (RR [95% CI], 1.47 (1.05-2.04]), and alcohol problem (RR [95% CI], 1.45 (1.04-2.02]) were found to have higher risks for bipolar disorder among depressive subjects. We found that a modified scoring of the MDQ (ignoring question on functional impairment and co-occurrence of symptoms) yielded a sensitivity of 0.68 and a specificity of 0.63 for bipolar diagnosis, whereas the figures were 0.29 and 0.77, respectively, with the standard MDQ scoring.

Conclusions: The results of this study clearly indicate that a high frequency of bipolar disorders in depressive patients who have never been diagnosed with bipolar disorders and clinical features indicating bipolarity could help to differentiate bipolar subjects from unipolar subjects. Adapting the standard scoring, the MDQ showed limited use for detecting bipolar disorder; however, if the scoring modification is adapted, the MDQ can offer tolerable sensitivity.

MeSH terms

  • Adult
  • Affect
  • Age of Onset
  • Alcoholism / psychology
  • Bipolar Disorder / classification
  • Bipolar Disorder / complications
  • Bipolar Disorder / diagnosis*
  • Depression, Postpartum / psychology
  • Depressive Disorder / complications*
  • Diagnostic and Statistical Manual of Mental Disorders
  • Female
  • Humans
  • Interview, Psychological
  • Male
  • Predictive Value of Tests
  • Recurrence
  • Sensitivity and Specificity
  • Suicide, Attempted / psychology
  • Surveys and Questionnaires