To help address this major challenge, the MedStar Health National Center for Human Factors in Healthcare team has developed an intuitive and interactive safety report visual dashboard. The dashboard allows safety leaders and analysts to quickly explore the data by seeing information at a department level while also seeing global trends across departments. The dashboard helps identify safety hazards more quickly.
We have also used data science techniques to develop algorithms that can more quickly analyze thousands of reports, including the free text, to identify specific themes and contributing factors pertaining to the safety issue. These algorithms rely on natural language processing and other techniques to make sense of the reports in a way that was never before possible.
The intuitive dashboard is being used by several safety analysts across MedStar Health and has also been shared with other healthcare systems as a model for how to make these report data more accessible.
Similarly, the algorithms are being applied directly to data at MedStar Health to identify important safety event themes and contributing factors. The algorithms have also been shared with other hospitals including the Children’s Hospital of Philadelphia and the Children's Hospital of Wisconsin.
Note: Click images to enlarge them.
Applied Clinical Informatics, “Using Active Learning to Identify Health Information Technology Related Patient Safety Events”
International Journal of Medical Informatics, “Integrating Natural Language Processing Expertise with Patient Safety Event Review Committees to Improve the Analysis of Medication Events”
Journal of Biomedical Informatics, “Identifying Health Information Technology Related Safety Event Reports from Patient Safety Event Report Databases”
Journal of the American Medical Informatics Association, “Connecting the Dots: Leveraging Visual Analytics to Make Sense of Patient Safety Event Reports”
Note: The patient information (name, time of admission, age, sex, etc.) presented in the dashboard examples on this page is fictitious.