Classification of Electronic Health Record-Related Patient Safety Incidents: Development and Validation Study

Area of Science:

• Health Informatics
• Patient Safety Research
• Sociotechnical Systems Analysis

Background:

• Implementation of health information technology (HIT) introduces new vulnerabilities in healthcare systems.
• A lack of standardized formats hinders the collection and analysis of technology-induced errors.
• Existing systems do not adequately address errors in mature electronic health record (EHR) environments.

Purpose of the Study:

• To develop a novel classification system for reporting patient safety incidents related to mature electronic health records (EHRs).
• To validate this classification system using incident data from a new EHR implementation.
• To evaluate the classification's characteristics and applicability in a real-world clinical setting.

Main Methods:

• The classification was developed using the Finnish Technology-Induced Error Risk Assessment Scale as a foundation.
• A multiprofessional team refined the classification through iterative consensus-building and testing.
• The developed classification was validated by analyzing 428 EHR-related incidents during a new EHR system's implementation phase, with interrater agreement assessed.

Main Results:

• EHR-related patient safety incidents increased fivefold during the new EHR implementation period compared to pre-implementation.
• Interface problems (22.5%) and usability issues (17.1%) were the most frequent error types identified.
• A significant portion of initial reports (14.8%) were rejected due to insufficient information, highlighting data quality challenges.

Conclusions:

• A new, validated classification system for EHR-related patient safety incidents in mature EHRs has been established.
• The high number of incidents during implementation underscores potential patient safety challenges.
• Error types evolve with the EHR development cycle, necessitating adaptive reporting and classification systems.