A pharmacoeconomic model of divalproex vs. lithium in the acute and prophylactic treatment of bipolar I disorder.

BACKGROUND Divalproex and lithium are the two most rigorously studied pharmacologic treatments for acute mania in bipolar I disorder in randomized, controlled trials. The differences between the drugs in their time course of onset, predictors of response, and side effects have potentially important pharmacoeconomic implications. METHOD Utilizing data from published studies, the University of Cincinnati Mania Project, and a consensus panel of psychiatrists, we developed a decision-analytic model to estimate the costs of treating patients with bipolar I disorder, acutely and prophylactically, for 1 year with divalproex or lithium. RESULTS In the overall group of patients with bipolar I disorder, initial treatment with divalproex led to costs that were 9% lower than costs for initial treatment with lithium. Cost savings associated with divalproex were greatest for patients with mixed mania and rapid cycling, whereas cost savings for patients with classic mania were greater for lithium. CONCLUSION According to the decision-analytic model developed in this study, divalproex, possibly because of a more rapid rate of antimanic activity associated with oral loading, is a less costly treatment than lithium in the acute and prophylactic treatment of patients with bipolar I disorder over 1 year.