Objective: To examine the impact of antipsychotic medication adherence on outcomes among individuals diagnosed with bipolar disorder. Methods: An administrative claims database for a commercially-insured population was used to identify patients with bipolar disorder who were newly initiating treatment with antipsychotics (January 2000–December 2006). Patients were included if they were aged between 18-64 years, had no diagnoses of dementia or schizophrenia, and were continuously insured from 6 months prior through 12 months post-index date (N=7769). Logistic stepwise regressions examined the association between achievement of adherence goals and patient outcomes (hospitalization or emergency room (ER) visit for any reason, mental-health related hospitalizations or ER visit), while controlling for demographic characteristics, type of bipolar disorder, general health, and comorbidities. Adherence was measured by the medication possession ratio (MPR). Results: The mean MPR was 41.65%, with 61.68% of individuals having an MPR of less than 0.50 and 78.67% having an MPR of less than 0.75. A significant reduction in the risk of hospitalization (odds ratio [OR=0.854]; 95% CI: 0.746–0.978) or an ER visit for any cause (OR=0.843; 95% CI: 0.744–0.955) was associated with an MPR of 0.75 or more. An MPR of 0.80 or more was associated with a significant reduction in the risk of a mental-health related hospitalization (OR=0.817; 95% CI: 0.699–0.954) while an MPR of 0.90 or more was associated with a significant reduction in the risk of a mental-health related ER visit (OR=0.705; 95% CI: 0.544–0.912). Incremental improvements in MPR from 0.75 to 0.95 were associated with corresponding reductions (20 to 30%) in the risk of any hospitalization or ER visit. Conclusion: As medication adherence improved, risk of hospitalization or ER visit declined, illustrating the link between adherence and better outcomes among patients with bipolar disorder.