Performance of a trigger tool for identifying adverse events in oncology.

Trigger tools are an increasingly important method to detect adverse events. A trigger is a finding identified through structured data that signals a potential adverse event that should be examined by a member of the health care team. In this retrospective cohort study of patients treated for breast, lung, or colorectal cancer, investigators tested a trigger tool with 76 components. After narrowing it to the 49 best-performing triggers, the tool had a positive predictive value of 0.48, meaning that nearly half of the triggers truly represented adverse events. Almost 20% of the triggers represented preventable events. The authors suggest that automation of their trigger tool could lead to real-time identification of adverse events in oncology. A past PSNet interview featured David Classen, one of the pioneers of the trigger tool methodology for detecting adverse events.