OBJECTIVES: Monotherapy with a mood stabilizer (MS) or second generation antipsychotic (SGA) is recommended as the first-line treatment for pediatric bipolar disorder (PBD).  The existing evidence regarding the relative effectiveness of MSs and SGAs for PBD is predominantly based on short-term studies and does not adequately address long-term effectiveness. This study compared adherence, persistence, and bipolar-related hospitalization of these treatments during a one-year observation period. METHODS: The 2003-2007 Medicaid Analytic eXtract data for four states were used. Bipolar children and adolescents (aged 6-18 years) initiating treatment with SGA or MS monotherapy were identified. Adherence was measured using medication possession ratio (MPR) and persistence was measured as time to medication discontinuation and time to augmentation. Survival Analyses was conducted to compare time to first bipolar-related hospitalization, time to discontinuation and time to augmentation between MS and SGA recipients during a one-year period after treatment initiation. Heckman’s Two-Step Selection Correction was used in all survival models to control for treatment selection bias. RESULTS: A total of 8424 PBD patients were identified. Prescription of SGAs (64.08%) was predominantly higher than that of MSs (35.92%). The most frequently prescribed SGA was risperidone, followed by quetiapine and aripiprazole. Divalproex sodium and oxcarbazepine were most frequently prescribed among MSs. 55% of the patients initiated on either of the therapeutic category were fully adherent. After correcting for selection bias, there was no statistically significant difference in the MPR, time to discontinuation and time to hospitalization between the two study groups. Patients initiating on SGAs took a longer time to augment (Hazard Ratio: 0.71; 95%CI: 0.57-0.88) with MSs as compared to those who initiated with MSs. CONCLUSIONS: Although SGAs were prescribed predominantly more than MSs, the two therapeutic classes were comparable in adherence and preventing bipolar related hospitalization. SGAs appeared to be slightly better than MSs in terms of time to augmentation.