Taxonomy development methods regarding patient safety in health sciences - A systematic review

Area of Science:

• Healthcare Informatics
• Clinical Data Analysis
• Patient Safety Research

Background:

• Automated analysis of clinical data necessitates standardized taxonomies.
• Existing reviews on taxonomy development methods in health sciences are limited.
• This systematic review addresses the scope of patient safety taxonomies and their development methodologies.

Purpose of the Study:

• To systematically review existing taxonomies related to patient safety.
• To describe the methods employed in health science taxonomy development.
• To identify strengths and limitations of current taxonomy development methods to guide future projects.

Main Methods:

• Systematic literature search of CINAHL, PubMed, Scopus, and Web of Science (Jan 2012–Apr 2023).
• Study selection by two authors using predefined inclusion/exclusion criteria and appraisal checklists.
• Inductive data analysis with narrative reporting of results.

Main Results:

• Thirteen studies focused primarily on adverse events and medication safety taxonomies, with limited coverage in specialized fields and IT.
• Inadequate reporting of taxonomy development methods was noted.
• Ten development phases were identified, with multimethod/multi-stakeholder approaches and piloting as strengths, while time consumption and small sample sizes in testing were limitations.

Conclusions:

• There is a need for new taxonomies in diverse medical specialties and health information technology to enhance patient safety.
• A structured methodology for taxonomy development, reporting, and appraisal is essential to improve quality.
• A proposed guide for taxonomy development requires further testing and validation.