This study evaluated the cost effectiveness of olanzapine compared with lithium as maintenance therapy for patients with bipolar I disorder (BP1) in the UK. A Markov model was developed to assess costs and outcomes from the perspective of the UK National Health Service over a 1-year period. Patients enter the model after stabilization of a manic episode and are then treated with olanzapine or lithium. Using the findings of a recent randomized clinical trial, the model considers the monthly risk of manic or depressive episodes and of dropping out from allocated therapy. health care resources associated with acute episodes were derived primarily from a recent UK chart review. Costs of maintenance therapy and monitoring were also considered. Key factors influencing cost effectiveness were identified and included in a stochastic sensitivity analysis. The model estimated that, compared to lithium, olanzapine significantly reduced the annual number of acute mood episodes per patient from 0.81 to 0.58 (difference -0.23; 95% CI: -0.34, -0.12). Per patient average annual care costs fell by 799 UK pounds (95% CI: – 1,824 UK pounds, 59 UK pounds) driven by reduced inpatient days–but the cost difference was not statistically significant. Sensitivity analysis found the results to be robust to plausible variation in the model’s parameters. The model estimated that using olanzapine instead of lithium as maintenance therapy for BP1 would significantly reduce the rate of acute mood events resulting in reduced hospital costs. Based on available evidence, there is a high likelihood that olanzapine would reduce costs of care compared to lithium.