The Emergency Department Trigger Tool: Validation and Testing to Optimize Yield.

Area of Science:

• Clinical Quality Improvement
• Patient Safety
• Healthcare Informatics

Background:

• Trigger tools are effective for adverse event (AE) detection in various clinical settings.
• An enhanced Emergency Department Trigger Tool (EDTT) was previously developed for high-yield AE detection.
• Validation of the EDTT in an independent sample and optimization of record selection are crucial for quality improvement.

Purpose of the Study:

• To validate the previously derived EDTT in an independent dataset.
• To compare different record selection approaches for optimizing AE detection yield.
• To identify triggers with persistent associations with AEs and develop a risk-scoring model.

Main Methods:

• Retrospective observational study of 92,859 adult patient records from an urban, academic ED.
• Validation of 30 triggers on 3,724 records; testing of 3 new candidate triggers.
• Bivariate and multivariable analyses to identify AE-associated triggers and develop risk scores; comparison of record selection strategies (random, trigger count, weighted trigger count).

Main Results:

• Twenty-nine triggers were confirmed to be associated with AEs, with an AE rate of 10% in records with triggers (vs. 1.1% without, p < 0.001).
• A multivariable model retained 12 triggers (ROC AUC 82%); weighted trigger counts yielded 18% for top 10% of visits and 38% for top 1%.
• Record selection methods did not affect the representativeness of detected AE types or severities.

Conclusions:

• The EDTT demonstrated high validity in trigger selection, yield, and AE representativeness in this single-site study.
• Weighted trigger count selection significantly outperformed random sampling and trigger count threshold approaches for AE detection yield.
• The EDTT is a promising, efficient, and high-yield tool for detecting all-cause harm to guide quality improvement in the ED.