Development and application of desiderata for automated clinical ordering

Area of Science:

• Clinical Informatics
• Health Systems Engineering
• Medical Decision Support

Background:

• Electronic Health Records (EHRs) offer potential for automating clinical orders to reduce clinician workload and enhance patient safety.
• A structured framework is needed to ensure clinical validity, transparency, and safety when automating orders.

Purpose of the Study:

• To develop and validate a framework of desiderata for assessing the appropriateness of automating clinical orders in EHRs.
• To demonstrate the operational value of this framework using a live health system dataset.

Main Methods:

• Focus group analyses and literature review to identify themes and develop desiderata.
• Clinician surveys to validate desiderata using use cases and Likert scales.
• Analysis of 1.4 million order-based alerts and 44.1 million orders using filtering rules and association rule mining.

Main Results:

• Identified 8 desiderata: logical consistency, data provenance, order transparency, context permanence, monitoring plans, trigger consistency, care team empowerment, and system accountability.
• Appropriate use cases scored significantly higher for appropriateness, cognitive support, and patient safety (P < .001).
• Operational analysis indicated potential time savings of 26.5 provider hours annually and identified opportunities for 8000 additional appropriate orders.

Conclusions:

• The developed desiderata align with clinician perceptions and offer a structured approach for evaluating automated orders.
• Automation of specific clinical orders shows potential for improving cognitive support while maintaining patient safety.
• Healthcare systems should utilize these desiderata and data mining for systematic identification and governance of automated orders.