Understanding Clinical Decision Support Failures in Pediatric Intensive Care Units via Applied Systems Safety...

Area of Science:

• Pediatric critical care medicine
• Human factors engineering
• Systems safety engineering

Background:

• Pediatric intensive care units (PICUs) present complex challenges for diagnosing critical illness, increasing vulnerability to decompensation and errors.
• Existing clinical decision support (CDS) tools in PICUs have shown limited success in improving clinician practices or demonstrating clinical benefit for diagnostic accuracy.

Purpose of the Study:

• To analyze workflow and failure modes in diagnosing and managing clinical decompensation within the PICU.
• To investigate the integration and impact of clinical decision support (CDS) tools using systems safety engineering and human factors principles.

Main Methods:

• Qualitative interviews and workflow mapping to identify communication and information access barriers.
• Immersive virtual reality (VR) systems testing using a digital twin environment to simulate and analyze staff interactions with CDS tools.
• Failure modes and effects analysis (FMEA) to pinpoint critical failure points.

Main Results:

• Workflow inefficiencies, communication barriers, and inadequate access to clinical information were identified as key challenges during critical events.
• Virtual reality simulations revealed how PICU staff interact with CDS tools and highlighted opportunities for better workflow integration.
• Specific critical failure modes were identified, offering targeted areas for intervention to enhance care delivery.

Conclusions:

• A systems safety and human factors approach, combined with VR and digital twin technology, provided valuable insights into diagnostic errors in the PICU.
• This methodology identified key failure modes and potential interventions to improve diagnostic accuracy and patient care in a quaternary referral center PICU.