Beyond a single index of mania symptoms: structure and validity of subdimensions.

BACKGROUND DSM-5 portrays mania as unitary despite evidence of distinct symptom clusters. Studies investigating the structure of mania have been inconsistent, in part because many relied on instruments not designed for this question. The present work used a clinical interview designed for structural analysis in order to identify and validate subdimensions specific to mania symptoms. METHODS Psychiatric outpatients (N=422) and undergraduates with a history of mental health treatment (N=306) were interviewed with a comprehensive measure of mood and anxiety that included 24 manic symptoms. Patients completed additional measures of symptoms and functioning, and a semi-structured diagnostic interview. RESULTS A 4-factor model of mania replicated across independent samples and was superior in fit to competing models, including the unidimensional model implied whenever researchers or clinicians use only a total score for mania. The factors were only moderately correlated, and three of the four ("Irritability" was the exception) showed a strong criterion, convergent and discriminant validity, suggesting they are specific to mania. Subdimensions showed distinct and meaningful associations with functioning. LIMITATIONS Symptoms of psychosis and depression are important features of manic episodes, but were not included in the present study since they lack specificity to mania. CONCLUSIONS Mania is multifaceted. At least three subdimensions specific to mania were identified ("Euphoric Activation," "Hyperactive Cognition" and "Reckless Overconfidence"). Use of subdimensions, in addition to overall mania severity, may enhance the ability of studies to detect meaningful biological correlates of bipolar disorder. Moreover, their different associations with functioning suggest assessing subdimensions has clinical utility as well.