Big data in anaesthesia: a narrative, nonsystematic review.

Area of Science:

• Anesthesiology
• Health Informatics
• Data Science

Background:

• Increasing data generation from anesthesia information management systems (AIMS) offers potential for scientific advancement.
• Lack of data integration hinders the effective use of routinely collected anesthesia data for improving patient care.
• Electronic health records (EHRs) have defined integration levels (e.g., NHS model) applicable to anesthesia practice.

Purpose of the Study:

• To review the utility of routinely collected data for enhancing care quality.
• To define EHR integration levels and illustrate their application in anesthesia.
• To explore data utilization strategies at micro, meso, and macro system levels.

Main Methods:

• Narrative review of literature on data integration in anesthesia.
• Definition and illustration of the six levels of EHR integration.
• Description of a custom AIMS solution and its integration opportunities.
• Discussion of research prospects and global opportunities, including in low-income countries.

Main Results:

• Routine anesthesia data, when integrated, can significantly improve patient care quality.
• A homemade AIMS solution demonstrated potential for enhanced data utilization across different system levels.
• Developing a core dataset for peri-operative research can facilitate large-scale EHR data integration.

Conclusions:

• Integrating AIMS data into EHRs is essential for continuous practice re-evaluation and optimal patient care.
• Training anesthesiologists in data science and AI is critical for advancing peri-operative medicine.
• Consideration of the ecological impact of data centers and global data sharing are important future directions.