OBJECTIVES: To assess the differential impact of treatment options for bipolar disorders on BMI in children and adolescents. METHODS: A retrospective study was conducted using the General Electric EMR database from 1995 to 2010 to identify bipolar individuals’ age 18 years or less who were prescribed atypical antipsychotics (AT), mood stabilizers (MS) or antidepressants (AD). The first bipolar diagnosis available in the data was defined as the index date. After excluding patients with bipolar diagnosis or medication prescription within 6 months pre-index and without at least one BMI measure pre-index and post-index, individuals were followed for up to 3, 6, 9 and 12 months if BMI data were available. Repeated measures models were applied to estimate the effect of treatment regimens on BMI after adjusting for the baseline BMI, socio-demographic factors, comorbidities and psychotherapy. The model helps account for the nesting effect of multiple BMI measures available for each individual. RESULTS: The cohort consisted of 2,299 individuals (mean age: 13.51±3.87 years). Children and adolescents on MS-monotherapy regimens showed a steady increase in BMI over time (3 months: 0.11 kg/m, 6 months: 0.09 kg/m, 9 months: 0.09 kg/m and 12 months: 0.089 kg/m, 6 months: 0.10 kg/m, 9 months: 0.07 kg/m, 12 months: 0.05 kg/m) at all time points compared to MS monotherapy. A further analysis on AT recipients showed that subjects with a greater baseline BMI and at younger age experienced smaller increases in BMI at all time points. CONCLUSIONS: Children and adolescents on all treatment options for bipolar disorder are significantly associated with an increase in BMI and with the greatest increase observed among individuals on AT regimens.