Affective Comorbidity in Psychiatrically Hospitalized Adolescents with Conduct Disorder or Oppositional Defiant Disorder: Should Conduct Disorder Be Treated with Mood Stabilizers?

Two hundred and twenty-three consecutively admitted adolescent inpatients were systematically assessed for conduct disorder (CD) and oppositional defiant disorder (ODD), using the Schedule for Affective Disorders and Schizophrenia in School-Aged Children (K-SADS) and DSM-III-R criteria. In this suburban private psychiatric hospital, 26% of adolescents met criteria for CD, and 12% met criteria for ODD. Whether they had CD or not, about two-thirds of these hospitalized adolescents had a mood disorder. In examining specific mood disorders associated with CD, bipolar disorder was significantly more common in patients with CD (25%) than in patients without CD (10%). No other mood disorder, including major depression or dysthymia, was more concentrated in the CD sample (compared to the non-CD sample). CD patients were also significantly more likely to have attention-deficient hyperactivity disorder (ADHD) and substance abuse, and significantly less likely to have anxiety disorders. In contrast, hospitalized adolescents with ODD had a much different pattern of comorbidity, with only a quarter having a mood disorder. In fact, adolescents with ODD were less likely to have a diagnosis of mood disorder than patients without ODD (27% versus 76%). No association was observed between ODD and the presence of ADHD, anxiety disorder, or substance abuse. These patterns of comorbidity cannot be generalized to adolescents in other settings or to children. However, pending direct pharmacological trials of efficacy, empirical trials of lithium or anticonvulsants in the adolescents with CD appear to be justifiable in our sample of psychiatrically hospitalized adolescents, since 25% appeared to have bipolar disorder. There was no diagnostic evidence to encourage attempts at antidepressant treatment of adolescents with CD, apart from the high rate of mood disorders generally seen in hospitalized adolescents. The pharmacological treatment of CD cannot be based purely on comorbidity findings, and must still be subjected to controlled studies of efficacy.