BACKGROUND: Isolated case reports have appeared implicating fluoxetine as a cause of impaired hemostatic function. The authors attempted to investigate this phenomenon. METHOD: The hemostatic function of 10 patients with DSM-III-R major depression was investigated before and after a trial of treatment with fluoxetine. The tests included international normalized ratio, partial thromboplastin time, factors II, V, VII, VIII:C, IX, X, XI, XII, fibrinogen, thrombin time, bleeding time, euglobulin lysis time, protein kinase C, antithrombin, platelet counts, D dimer, lupus inhibitor, and platelet sensitivity studies to the following agonists: adenosine diphosphate, epinephrine, collagen, and arachidonic acid. Hemostatic data on a patient with a drug-related bleeding episode are not available. RESULTS: No significant difference was found in any of the parameters tested. CONCLUSION: This study failed to demonstrate any compromised hemostatic function associated with fluoxetine therapy. However, the possibility of a type II statistical error exists. It is possible that bleeding events may occur as an idiosyncratic reaction to fluoxetine treatment.