EXPERT-SYSTEM DEVELOPMENT IN NURSING - IMPLICATIONS FOR CRITICAL CARE NURSING PRACTICE

Objective: To obtain information about how highly experienced critical care nurses reason to plan care and make decisions about a critically ill unstable patient, and to determine the usefulness of this information for expert system development. Design: Descriptive, using think-aloud technique and protocol analysis. Setting: Laboratory. Patient: A simulated patient case whose condition deteriorated over a 12-hour shift. The case depicted an elderly female with congestive heart failure and atrial fibrillation with rapid ventricular response. Results: Protocol analysis revealed the information (data) that subjects used and how they structured that information to plan care and make decisions. Examination of subjects' reasoning processes allowed the investigators to identify ''if-then'' rules that could be used in expert system design. Conclusions: The reasoning processes identified would assist in expert system development. An expert system designed to represent experienced critical care nurses' knowledge and reasoning processes would preserve that expertise in a computer system that could then be used to assist less experienced nurses to improve their reasoning skills and strategies.