Impact of an Alert-Based Inpatient Clinical Decision Support Tool to Prevent Drug-Induced Long QT Syndrome:...

Area of Science:

• Clinical Informatics
• Cardiology
• Pharmacology

Background:

• Clinical decision support (CDS) tools are integrated into electronic health records to alert clinicians about potential drug-induced QT prolongation (diLQTS).
• The effectiveness of these systems often relies on accurately identifying high-risk patients, but risk prediction alone may not guarantee improved outcomes.

Purpose of the Study:

• To evaluate the impact of a system-wide, alert-based inpatient CDS tool designed to prevent diLQTS.
• To assess the effect of the CDS tool across 10 known QT-prolonging medications on patient outcomes.

Main Methods:

• A comparative analysis of 178,097 hospitalizations (102,847 patients) before and after CDS implementation.
• Outcomes including diLQTS risk, hospitalization duration, and mortality were compared, with adjustments for patient demographics, comorbidities, and laboratory values.

Main Results:

• The CDS successfully identified high-risk patients for diLQTS (OR 2.28).
• No significant impact was observed on diLQTS risk or mortality based on provider responses to alerts.
• The overall risk of diLQTS did not significantly change post-implementation, while mortality rates increased.

Conclusions:

• Alert-based CDS systems, while effective at identifying high-risk patients for diLQTS, did not reduce the actual risk of diLQTS.
• System-wide implementation of CDS tools requires comprehensive evaluation beyond risk identification to ensure improved clinical outcomes.