In a recent study of Medicare patients with atrial fibrillation discharged from the hospital, Whittle et al1 reported that approximately one third of the patients for whom it appeared appropriate were not treated with warfarin sodium. Their evaluation demonstrated widely varying rates of use of warfarin for atrial fibrillation among the 5 hospitals sampled.
In a recent project initiated subsequent to the unpublished study referred to in the article,1 we evaluated the use of both warfarin and aspirin in Medicare patients with atrial fibrillation who were discharged alive from the hospital. We retrospectively reviewed a 50% sample of patients aged 65 to 85 years who were discharged with a principal or secondary diagnosis of atrial fibrillation (International Classification of Diseases, NinthRevision, Clinical Modification, diagnosis code 427.31) between July 1, 1995, and December 31, 1995, from 12 hospitals in the state of Oklahoma. The methods of our project were very