Detecting adverse events through automated global triggers

Increased adoption of electronic health records (EHRs) creates a vast knowledge base including large amounts of data on the course of patient treatment, including adverse events. Comprehensive insight into these harmful events can lead to interventions that improve quality of care and patient safety.

The Institute for Healthcare Improvement (IHI) Global Trigger Tool (GTT) for Measuring Adverse Events is designed to detect triggers indicating the possibility of an adverse event during the treatment process. Many of these events can be avoided by detecting patterns that allow for the design and use of safer procedures.

However, the GTT is primarily based on manual reviews. Since screening large volumes of natural language, unstructured text is laborious, automated methods are highly desirable. In a pilot study of patients, a European hospital identified a subset of triggers in the GTT and used text mining to automate the discovery of adverse events. They achieved very high correlation between adverse events detected through automation, compared with those detected manually. As a result, they were able to define the triggers most associated with avoidable adverse events.

This hospital worked with CGI to develop a continuous automated information system that analyzes all EHRs daily and produces a database of detected triggers. Based on the GTT, the system retrieves all text data for a specified medical specialty from the hospital's EHRs. The results are brought to clinicians in a secure, cloud-hosted dashboard allowing them to review those patient records that have presented the designated triggers. Documents containing the triggers statistically associated with avoidable adverse events are then analyzed to determine patterns in the treatment and the root causes leading to those events.

CGI’s Patient Safety Service provides a comprehensive electronic dashboard of confirmed adverse events affecting patients in the course of treatment. Improved patient safety can be promoted and the impacts of interventions can be followed up over time. The Patient Safety Service dashboard is designed to deliver and support outputs and outcomes such as:

Stage 1

  • Automatic identification of potential adverse events
  • Supportive analysis for remediating areas of concern
  • Reduced resources for manual adverse event and preventable medical error reporting
  • Reduced adverse events
  • Reduced costs with respect to secondary complications
  • Reduce costs with respect to quality and risk management

Stage 2

  • Automated information collation for aggregate use
  • Automated tooling integrating policy changes in support of patient safety
  • Integration to best practice protocol, clinical guidelines and cost conformance tools
  • Automated identification of adverse events in real time

Stage 3

  • Automated prescriptive analytics dedicated to preventing adverse events and medical error before they occur
  • Increased sustainability of healthcare

Advanced analytics capabilities can be applied to patient safety program data offering great promise for identifying and catching adverse events before they lead to patient harm. This innovative service provides the foundation for future proactive prevention of adverse events, helping hospitals to pursue the IHI “Triple Aim” framework to: 1) improve the patient experience of care, 2) improve the health of populations, and 3) reduce the per capita cost of healthcare. 

Contact us to learn more about Patient Safety Services.