Medical error is one of the leading causes of death in the US. The study and reduction of medical errors have become a major concern in healthcare today. It is believed that medical error reporting systems could be a good resource to share and to learn from errors if medical error data are collected in a properly structured format and are useful for the detection of patterns, discovery of underlying factors, and generation of solutions. Effectively gathering information from previous lessons and timely informing the subsequent action are the two major goals for the design, development and utilization of such a system. The Common Formats (CFs) suggested by AHRQ tend to unify the future reporting format, which holds promise in improving data consistency and reducing unsafe conditions through lessons learned. However, effective gathering medical incident data does not merely rely on a unified structure. To be able to learn from previous lessons, it heavily depends upon the quality reports and learning features offered by systems. Medical incident data are always the key components and invaluable assets in patient safety research. The long term goal of the project is to understand the occurrence and causes of medical incidents in real practice and to develop interventions based on collection of incident reports to minimize the recurrence of similar incidents that have been reported. The objective of this application is to improve the utilization f voluntary reporting systems that each healthcare institution has been put in use by developing a learning toolkit that can systematically collect and analyze incident reports, automatically link historical reports with WebM&M, the highest quality of voluntary reports and expert reviews in patient safety. As moving toward CFs, the researchers propose a user-centered, learning-supportive, and ontological approach that will help reporters generate complete and accurate reports through user-friendly guidance and offer timely comments and relevant peer reviews through educational tools during and after incident reporting. The researchers employ a case-based reasoning and natural language processing techniques to demonstrate the feasibility and effectiveness of the knowledge-based toolkit which helps reporters improve the communication about patient safety through clear working definitions and advance training that builds knowledge about the safety culture and then provides continuing education through the system. The project holds promise in revolutionizing the design of voluntary medical incident reporting systems from an incident data repository to an advanced resource promoting complete and accurate incident reporting and learning toward a just and learning culture.

Public Health Relevance

Timely reporting and effective learning from medical incidents is considered an effective way in developing strategies for reducing medical errors. Utilizing an innovative a user-centered, learning-supportive, and ontological approach combining with case-based reasoning and natural language processing techniques, we propose to develop a knowledgebase and learning toolkit that can systematically collect and analyze incident reports, linking historical reports with WebM&M, the highest quality of voluntary reports and expert reviews on patient safety. We envision that the innovative approach will facilitate timely, quality reporting and learning from the incidents and ultimately cultivating a just and learning culture of patient safety.

Agency
National Institute of Health (NIH)
Institute
Agency for Healthcare Research and Quality (AHRQ)
Type
Research Project (R01)
Project #
5R01HS022895-04
Application #
9352770
Study Section
Health Care Technology and Decision Science (HTDS)
Program Officer
Eldridge, Noel
Project Start
2014-09-30
Project End
2019-09-29
Budget Start
2017-09-30
Budget End
2018-09-29
Support Year
4
Fiscal Year
2017
Total Cost
Indirect Cost
Name
University of Texas Health Science Center Houston
Department
Type
Sch Allied Health Professions
DUNS #
800771594
City
Houston
State
TX
Country
United States
Zip Code
77030
Zhou, Sicheng; Kang, Hong; Yao, Bin et al. (2018) Unveiling Originated Stages of Medication Errors: An Automated Pipeline Approach. Stud Health Technol Inform 250:182-186
Wang, Ju; Ali, Eliza; Gong, Yang (2018) An Information Enhanced Framework for Reporting Medication Events. Stud Health Technol Inform 250:169-173
Yao, Bin; Kang, Hong; Wang, Ju et al. (2018) Exploring Health Information Technology Events from FDA MAUDE Database. Stud Health Technol Inform 250:187-191
Kang, Hong; Yu, Zhiguo; Gong, Yang (2017) Initializing and Growing a Database of Health Information Technology (HIT) Events by Using TF-IDF and Biterm Topic Modeling. AMIA Annu Symp Proc 2017:1024-1033
Wang, Ju; Gong, Yang (2017) Potential of Decision Support in Preventing Pressure Ulcers in Hospitals. Stud Health Technol Inform 241:15-20
Kang, Hong; Gong, Yang (2017) Developing a similarity searching module for patient safety event reporting system using semantic similarity measures. BMC Med Inform Decis Mak 17:75
Zhou, Sicheng; Kang, Hong; Gong, Yang (2017) Design a Learning-Oriented Fall Event Reporting System Based on Kirkpatrick Model. Stud Health Technol Inform 245:828-832
Yao, Bin; Kang, Hong; Miao, Qi et al. (2017) Leveraging Event Reporting Through Knowledge Support: A Knowledge-Based Approach to Promoting Patient Fall Prevention. Stud Health Technol Inform 245:973-977
Kang, Hong; Gong, Yang (2017) Design of a User-Centered Voluntary Reporting System for Patient Safety Events. Stud Health Technol Inform 245:733-737
Kang, Hong; Wang, Frank; Zhou, Sicheng et al. (2017) Identifying and Synchronizing Health Information Technology (HIT) Events from FDA Medical Device Reports. Stud Health Technol Inform 245:1048-1052

Showing the most recent 10 out of 22 publications