作者: C.Seth Landefeld , Philip A. Anderson , Lawrence T. Goodnough , Thomas W. Moir , David L. Hom
DOI: 10.1016/0895-4356(89)90066-8
关键词:
摘要: Reports of bleeding complications medical therapy should be based on valid methods classification, but the reproducibility existing has not been tested. Therefore, we prospectively studied three to classify severity bleeding: a purely subjective implicit method, previously published explicit method using brief criteria, and index, which is new detailed criteria about amount, rate, consequences bleeding. Three physicians independently reviewed abstracts 168 patients treated with anticoagulants. The proportion cases classified as major varied widely when was used (2, 14 39%), less old (28, 40 47%), at all index (20, 20 20%). Intraobserver agreement excellent for both (kappa greater than or equal 0.95). However, interobserver better = 0.87) 0.69) 0.39). We conclude that classification depends used. In comparison older methods, highly reproducible tested more determine whether it can applied burgeoning clinical research in anticoagulation thrombolysis.