Assessing Health Care Professionals' Perceptions of a New System in Clinical Workflows: Systems Engineering...

Area of Science:

• Healthcare Informatics
• Artificial Intelligence in Medicine
• Patient Safety

Background:

• AI-enabled clinical decision support systems (CDSSs) are increasingly integrated into electronic health records (EHRs).
• Introduction of CDSSs can disrupt workflows and raise safety concerns, especially in critical areas like surgical transfusion.
• Limited qualitative data exists on frontline professionals' anticipations of these systems before deployment.

Purpose of the Study:

• To qualitatively examine anticipated clinical, organizational, and workflow implications of implementing a specific AI-enabled CDSS (pMSBOS-TS) for surgical blood ordering.
• To gather insights before large-scale deployment of the personalized Maximum Surgical Blood Order Schedule-Thoracic Surgery (pMSBOS-TS).

Main Methods:

• Consensual qualitative study involving 14 multidisciplinary healthcare professionals.
• Two semistructured focus groups conducted after a pilot session.
• Analysis using the Systems Engineering Initiative for Patient Safety (SEIPS) framework, focusing on People, Environment, Tools, and Tasks.
• Task- and workflow-based analyses of transfusion processes and member checking for validity.

Main Results:

• Participants foresee reduced variation in blood ordering if the AI CDSS is reliable and integrated into EHR workflows.
• Concerns include increased verification burden, system limitations in unusual cases, and communication issues between units and the blood bank.
• Organizational culture, governance, and logistics are critical for determining if the system reduces or increases workload and waste.

Conclusions:

• Successful adoption of AI-enabled transfusion CDSSs hinges on sociotechnical readiness, including user trust, workflow fit, and organizational support, beyond just predictive performance.
• Findings offer practice-based insights for staged implementation, training, and governance to safely integrate predictive transfusion CDSSs into surgical workflows.