The problem of human error in healthcare is well documented. Other domains (such as transportation and energy) have used techniques and methods from the engineering disciplines to analyze and ultimately reduce instances of error. In New York State (NYS), the Department of Health (DOH) requires the use of one such method, Root Cause Analysis (RCA), in investigating and reporting the occurrence of and response to sentinel events by hospitals. Despite the use of the RCA technique in an attempt to identify and mitigate the root causes of error, the problem of human error contributing to sentinel events persists. Experts in Human Reliability Analysis (HRA) contend that human performance is too complex to be represented by models used in engineering systems reliability. A method to analyze human erroneous actions was created that considers operator context and control based on a model of human cognition. The technique, called the Cognitive Reliability and Error Analysis Method (CREAM), was used to reanalyze 58 sentinel event cases Rochester General Hospital previously analyzed using RCA. Despite serious data limitations, our results reveal an apparent gap between RCA and CREAM analyses. We suggest that the gap highlights incomplete RCA that minimizes or does not appreciate the role that organizational factors play in contributing to the sentinel events. Due to data limitations, we cannot identify specific interventions to mitigate risk for sentinel events. However, we make several recommendations for improving the RCA process at Rochester General Hospital in an effort to improve the validity of the analyses data for further study.

Publication Date



Technical report prepared for RIT-RGHS Alliance

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works in February 2014.

Document Type

Technical Report

Department, Program, or Center

Psychology (CLA)


RIT – Main Campus