Dissecting Patient Safety Events through Data Science
Medical errors are the 3rd leading cause of death in the US. Patient Safety Event (PSE) reporting has been a national priority since the IOM report “To Err is Human” in 2000. However, electronically collected reports are poor in interpretability and there is an apparent gap in translating the report to actionable knowledge for safety improvement. We developed (1) a prototype upper-class ontology for encoding terminologies used in PSE, (2) a machine-learning based text mining pipeline for categorizing PSE by error types, severity levels, contributing factors, and other clinically valuable information, (3) theories, methods, and clinical implementation strategies for applying these informatics tools in real-world Root Cause Analysis (RCA) and clinical decision support.
- Liang, C., Zhou, S., Yao, B., Hood, D., & Gong, Y. (2019). Toward systems-centered analysis of patient safety events: improving root cause analysis by optimized incident classification and information presentation. International Journal of Medical Informatics, 104054. full text
- Liang, C., Miao, Q., Kang, H., Vogelsmeier, A., Hilmas, T., Wang, J., & Gong, Y. (2019). Leveraging Patient Safety Research: Efforts Made Fifteen Years Since To Err Is Human. Studies in health technology and informatics, 264, 983-987. full text
- Liang, C., & Gong, Y. (2017). Predicting Harm Scores from Patient Safety Event Reports. In MedInfo (pp. 1075-1079). full text
- Liang, C., & Gong, Y. (2017). Automated Classification of Multi-Labeled Patient Safety Reports: A Shift from Quantity to Quality Measure. In MedInfo (pp. 1070-1074). full text
- Liang, C., & Gong, Y. (2016). Knowledge representation in patient safety reporting: an ontological approach. Journal of Data and Information Science, 1(2), 75-91. full text