Methodological Approaches for Analyzing Medication Error Reports in Patient Safety Reporting Systems: A Scoping Review.

Background: Medication errors (MEs) pose risks to patient safety, resulting in substantial economic costs. To enhance patient safety and learning from incidents, health care and pharmacovigilance organizations systematically collect ME data through reporting systems. Despite the growing literature on MEs in reporting systems, an overview of methods used to analyze them is lacking. The authors aimed to identify, explore, and map available literature on methods used to analyze MEs in reporting systems.

Methods: The review was based on Joanna Briggs Institute's methodology. The authors systematically searched electronic databases Embase, Medline, CINAHL, Cochrane Central, and other sources (Google Scholar, health care safety and pharmacovigilance centers' websites). Literature published from January 2017 to December 2023 was screened and extracted by two independent researchers.

Results: Among the 59 extracted publications, analyses most often focused on MEs occurring in hospitals (57.6%), included both adult and pediatric patients (79.7%), and used national patent safety monitoring systems as a source (69.5%). We identified quantitative (39.0%), qualitative (11.9%), mixed methods (37.3%), and advanced computerized methods (11.9%). Descriptive quantitative analyses for categorized data were common; however, disproportionality analysis constituted a newer approach to address issues with reporting bias. Free-text data were commonly managed by content analysis, while mixed methods analyzed both categorized and free-text data. In addition, text mining, natural language processing, and artificial intelligence were used in more recent studies.

Conclusion: This scoping review uncovered a notable span and diversity in methodologies. Future research should assess the use, applicability, and effectiveness of newer methods such as disproportionality analysis and advanced computerized techniques.