Inexperienced physicians have a higher rate of human error incidents than their more experienced colleagues and human errors far outnumber equipment malfunctions in the ICU. Physicians relying on intuitive thinking may be unable to comprehend the aggregate meaning of complicated ICU data bases and hence be unable to formulate the therapeutic action. Many investigators have found that performance of physicians and other health care providers is improved when they use protocols. Computer protocols for part of the management of mechanical ventilation (respiratory evaluation, oxygenation, weaning and extubation) in patients with adult respiratory distress syndrome (ARDS) have already ben developed and clinically validated at the LDS Hospital. These protocols have controlled decision making 94% of the 24 hour day when used for over 30,000 hours in 101 ARDS patients. 52 of these ARDS patients met extra corporeal membrane oxygenation (ECMO) criteria and their survival was 41% (4X the expected 9% from historical data (p<0.0002)). These computer protocol results clearly established the feasibility of controlling the therapy of severely ill patients. Our long term goal is to generate and test the efficacy of computer protocols for the management of critically ill patients. This specific project will focus on developing and testing the efficacy and generalizability of protocols for one of the demanding aspects of critical care. the management of mechanical ventilation. We will use the existing LDS Hospital HELP system infrastructure to develop and clinically test the protocols for management of mechanical ventilation (respiratory evaluation, oxygenation, ventilation, weaning and extubation). These protocols will be implemented using HELP Frames which conform to the Arden standard for medical knowledge representation. Once the protocol logic has been shown to be sound we will transfer the protocols to a portable PC in order to conduct a two year prospective randomized clinical trial to test efficacy of computerized protocols in 300 patients with ARDS at Charles R. Drew Univ/Martin Luther King, Jr. (CRDU/MLK) Hospital in the WATS district of Los Angeles, CA (H-0: There is no difference in efficacy between protocol and non-protocol controlled critical care). CRDU/MLK Hospital does not have a hospital information system, has not participated in the development of the protocols, and has not used these protocols previously. We will define efficacy using a hierarchal four level approach; Efficacy->a) Survival, b) Morbidity, c) Incidence and severity of barotrauma, d) Resource consumption. We will evaluate protocol performance during the randomized trial at CRDS/MLK Hospital in order to assess the generalizability of the protocols by examining; 1) Percent of total time in the trial during which protocols controlled patient care. 2) Number of protocol instructions which were not followed. 3) Number of objections to protocol logic which, based on medical evidence, forced a change in logic.
Showing the most recent 10 out of 17 publications
