Accuracy of decisions to withdraw therapy in critically ill patients: clinical judgment versus a computer model.

摘要: Two clinicians and the nursing sisters working in ICU evaluated chance of survival patients every day. Patients were assessed either as "outcome unknown or will die." These predictions compared with those made by computerized trend analysis daily acute physiology chronic health evaluation (APACHE II) scores corrected for presence duration major organ system failure. The not acted upon during study. Comparing actual hospital outcome, doctors nurses had a false-positive diagnosis rate dying between 7.7% 16.7%, while there no false computer model. predicted to die completely identical computer. Predictions that confirmed sensitivity 20% death.