Family history validation of a definition of mixed depression

Purpose: The study aim was to test different definitions of mixed depression, defined as a depression with concurrent hypomanic symptoms. Methods: Consecutive 245 non-tertiary care outpatients with bipolar II disorder (BP-II) and 189 non-tertiary care outpatients with major depressive disorder (MDD) were interviewed (off psychoactive drugs) using the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Axis I Disorders-Clinician Version, Hypomania Interview Guide (HIG), and Family History Screen when presenting for major depressive episode (MDE) treatment. Intra-MDE hypomanic symptoms were systematically assessed. Mixed depression was defined as an MDE with concurrent hypomanic symptoms. Receiver operating characteristic (ROC) analysis and multivariate analysis were used to test different definitions of mixed depression (dimensional and categorical ones). Factor analysis was also used. Bipolar family history was the validator. Findings: Bipolar II disorder, vs MDD, had significantly more intra-MDE hypomanic symptoms (racing/crowded thoughts, irritable mood, psychomotor agitation, more talkativeness, and increased goal-directed and risky activities). Major depressive episode plus 3 or more hypomanic symptoms was present in 68.7% of BP-II and 42.3% of MDD. A "motor activation" factor, including psychomotor agitation and talkativeness, and a "mental activation" factor including racing/crowded thoughts were found. Different definitions (dimensional and categorical ones) of mixed depression were tested vs bipolar family history as validator (ie, MDE plus more than 1, 2, 3, and 4 concurrent hypomanic symptoms, MDE plus psychomotor agitation, MDE plus racing thoughts). Major depressive episode plus more than I hypomanic symptom had the highest sensitivity but the lowest specificity. Instead, MDE plus more than 4 hypomanic symptoms had the lowest sensitivity and the highest specificity. The better-balanced combination of sensitivity and specificity was shown by MDE plus more than 2 hypomanic symptoms. The same definition also showed the highest ROC area value. Multivariate regression of bipolar family history vs different mixed depression definitions found that the only strong and significant predictor was MDE plus more than 2 hypomanic symptoms. A dose-response relationship was found between the number of hypomanic symptoms during MDE and the bipolar family history loading. Conclusions: Mixed depression (MDE plus 3 or more hypomanic symptoms) was common in BP-II and MDD. A dimensional definition based on 3 or more hypomanic symptoms during depression was the most supported by using bipolar family history as validator. The study of mixed depression may be important for its possible impact on treatment (antidepressants could increase hypomanic symptoms, and mood stabilizers and antipsychotics could control hypomanic symptoms during antidepressant treatment). (c) 2005 Elsevier Inc. All rights reserved.