부산대학교 도서관

The actual and preferred methods of continuing medical education. (CME) on diabetes were determined from a telephone survey of a stratified random sample of over 600 primary care physicians in Pennsylvania. A response rate of 73% was obtained. Journal articles were identified as the leadingsource of information that affected practice behavior related to diabetes. However, no single journal was read by a majority ofphysicians, and information provided in diabetes journals reached fewer than 10% of primary care physicians. Local educational programs and programs offered by hospitals were the preferred choices for continuing medical education. When reported patterns of care for diabetic patients were compared with published standards of care, no statistical differences were found between physicians who had and physicians who had not attended a CME course. To be most effective, CME courses should be intensive and may need to be presented locally to reach the widest possible audience.