Applying Computational Ethnography to Examine Nurses' Workflow Within Electronic Health Records

Area of Science:

• Health Informatics
• Nursing Informatics
• Computational Social Science

Background:

• Electronic Health Records (EHRs) are critical in modern healthcare, but their workflow analysis is often inflexible.
• Nursing research requires advanced methodologies to understand complex EHR interactions.
• Computational ethnography presents a novel approach to supplement existing EHR analysis techniques.

Purpose of the Study:

• To describe computational ethnography as a methodology for EHR workflow analysis in nursing.
• To highlight the relevance of computational ethnography for nursing research.
• To explore data mining techniques for analyzing nurses' EHR navigation patterns.

Main Methods:

• Utilizing audit logs as a data source for computational ethnography.
• Applying data mining techniques: Sequential Pattern Mining (SPM) and Markov Chain Analysis (MCA).
• Analyzing nurses' workflow and navigational patterns within EHR systems.

Main Results:

• Demonstrated hypothetical examples of SPM identifying frequent workflow sequences.
• Showcased MCA for calculating transition probabilities between EHR screens.
• Highlighted the ability of these methods to uncover global navigational patterns.

Conclusions:

• Computational ethnography and data mining offer robust methods for EHR workflow analysis.
• These techniques can inform EHR optimization by identifying navigation patterns and predicting future use.
• The findings support the redesign of EHRs to improve nursing efficiency and user experience.