January 1990

Physician Variation in Anticoagulating Patients With Atrial Fibrillation

Author Affiliations

From the Department of Community and Family Medicine, Dartmouth Medical School (Drs Chang, Kirk, and Wasson), and the Department of Anesthesiology, Mary Hitchcock Medical Center (Dr Deroo), Hanover, NH; and the Department of Internal Medicine, Veterans Administration Medical Center, White River Junction, Vt (Drs Bell and Wasson).

Arch Intern Med. 1990;150(1):83-86. doi:10.1001/archinte.1990.00390130089012

• We investigated variations in the oral anticoagulant treatment of atrial fibrillation by physicians in three specialties: family physicians (or general practitioners), general internists, and cardiologists. Results showed general agreement in the anticoagulation decision regarding patients with either mitral valve disease or a history of chronic alcohol abuse, but substantial disagreement in other categories of patients. Estimations of the risk of embolization and risk of hemorrhage differed widely among all physicians, cardiologists generally rating the embolization risks lower than the other physicians. A physician's treatment decision was strongly related to the relative risk of embolism vs hemorrhage derived for each case. A relationship between physician specialty and treatment decision was also demonstrated, with cardiologists least likely, and family practitioners most likely, to institute anticoagulation in nonrheumatic patients with atrial fibrillation. The reason for this variation appears to be differences in the estimated risk of systemic embolism.

(Arch Intern Med. 1990;150:81-84)