Medical errors have been shown to be the third leading cause of death in the United States. The Institute of Medicine and several state legislatures have recommended the use of patient safety event reporting systems (PSRS) to better understand and improve safety hazards. A patient safety event (PSE) report generally consists of both structured and unstructured data elements. Structured data are pre-defined, fixed fields that solicit specific information about the event. The unstructured data fields generally include a free text field where the reporter can enter a text description of the event. The text descriptions are often a rich data source in that the reporter is not constrained to limited categories or selection options and is able to freely describe the details of the event. The goal of this project is to develop novel statistical methods to analyze unstructured text like patient safety event reports arising in healthcare, which can lead to significant improvements to patient safety and enable timely intervention strategies. We address three problems: (a) Building realistic and meaningful baseline models for near misses, and detecting systematic deterioration of adverse outcomes relative to such baselines; (b) Understanding critical factors that lead to near misses & quantifying severity of outcomes; and (c) Identifying document groups of interest. We will use novel statistical approaches that combine Natural Language Processing with Statistical Process Monitoring, Statistical Networks Analysis, and Spatio-temporal Modeling to build a generalizable toolbox that can address these issues in healthcare. We will also release open source software via R packages & GitHub, which will enable healthcare staff and researchers to execute our methods on their datasets. The COVID-19 pandemic has resulted in increased patient volumes and increased patient acuity, leading to an excessive burden on many healthcare facilities across the United States. This greatly increases the risk of patient safety consequences arising from malfunctioning medical equipment or adverse reaction to medication. To ensure patient safety and the highest quality of healthcare during this crisis, we need a rapid response system to model and analyze COVID-specific safety issues at scale, and quickly disseminate the results to healthcare facilities, so that these risks can be mitigated at the point of care. In this supplement, we propose to do this by (a) mining public databases and EHRs to identify devices/medication being used for treating COVID and (b) applying our methods (based on NLP, SPC, and SPM) to understand risks associated with these items. This information will be disseminated nationally to all healthcare facilities so that it can be integrated into the EHR at the point of care to alert clinicians.
Estimates of preventable adverse events in healthcare are staggering, and the risk is particularly high for COVID patients due to the rapidly increasing burden on healthcare facilities. Using our algorithms to identify temporal trends and analyze free text narratives from reporting systems can ensure the safety and quality of care for COVID patients by exposing and mitigating possible weaknesses in the care process.
