Visualization of aggregate perioperative data improves anesthesia case planning: A randomized, cross-over trial

Area of Science:

• Medical Informatics
• Anesthesiology
• Surgical Education

Background:

• Managing surgical procedure variation is challenging due to privacy and logistical constraints.
• Reviewing prior cases to learn from past practices is often not feasible.
• A system to extract key data from anesthesia records can bridge this gap.

Purpose of the Study:

• To determine if a data visualization system improves case planning performance for anesthesia residents.
• To assess the impact of a tool for reviewing de-identified anesthesia records.
• To evaluate the feasibility of aggregating key anesthetic information.

Main Methods:

• A randomized, cross-over trial was conducted at Vanderbilt University Medical Center.
• A web-based data visualization tool was developed for de-identified anesthesia records.
• First-year anesthesia residents performed simulated case planning tasks, with performance scored by an algorithm.

Main Results:

• The tool's utilization was associated with an average performance score improvement of 0.120 points (p < 0.001).
• A significant "learning effect" improved scores with each assessment (p < 0.001).
• Performance scores varied significantly across different case scenarios (p < 0.001).

Conclusions:

• A clinical data visualization system for anesthetic information is feasible.
• The tool modestly improved residents' simulated case planning performance.
• This approach aids in managing care variation by facilitating learning from prior practices.