Diagnostic characteristics of child bipolar I disorder: does the "Treatment of Early Age Mania (team)" sample generalize?

OBJECTIVE

To examine the representativeness of a randomized controlled trial (RCT) sample versus one obtained by consecutive new case ascertainment, for subjects with child bipolar I disorder.

METHOD: Subjects (N = 247) were outpatients who participated in either the National Institute of Mental Health-funded Phenomenology and Course of Pediatric Bipolar Disorders study or the Treatment of Early Age Mania (TEAM) study. Both studies required that subjects have current DSM-IV bipolar I disorder (manic or mixed phase) and a Children's Global Assessment Scale (CGAS) score <or=60. All subjects had elation and/or grandiosity. Subjects in the Phenomenology study were obtained from 1995 to 1998 by consecutive new case ascertainment from designated pediatric and psychiatric facilities. Subjects in the TEAM RCT were recruited from media and community sources between March 2003 and March 2005. Assessment instruments included the Washington University in St. Louis Kiddie Schedule for Affective Disorders and Schizophrenia, given separately to parents about their children and to children about themselves, and the CGAS. Logistic regression was used for comparisons.

RESULTS

The TEAM and Phenomenology groups were similar in age (10.4 [SD = 2.3], 10.9 [SD = 2.3] years, respectively) and other demography. Both had long current episode duration (4.8 [SD = 2.4], 3.2 [SD = 2.3] years) and low lifetime use of any mood stabilizer (23.6%, 35.0%). Many mania symptoms and ultradian rapid cycling, psychosis, and suicidality were significantly more prevalent in the RCT sample.

CONCLUSIONS

Generalization of the RCT sample was supported, because only 7.8% of Phenomenology subjects did not fit the RCT criteria. Nevertheless, because the RCT subjects were more severely ill, it is unclear if treatment findings from the RCT will be applicable to children with less severe mania.

CLINICAL TRIALS REGISTRATION: ClinicalTrials.gov identifier NCT00057681 (TEAM study).