Topic Analysis of UK Fitness to Practise Cases: What Lessons Can Be Learnt?

Area of Science:

• Healthcare professional regulation
• Medical informatics
• Computational linguistics

Background:

• Fitness to practise (FtP) impairment among UK healthcare professionals poses risks to patient safety, professional reputation, and careers.
• Publicly available FtP case documents offer learning opportunities but are time-consuming to review.
• Machine learning (ML) can streamline the analysis of large volumes of FtP case data across professions.

Purpose of the Study:

• To demonstrate how ML can facilitate the examination of FtP cases for UK dental, medical, nursing, and pharmacy professionals.
• To identify common and profession-specific themes in FtP impairment cases using ML.

Main Methods:

• Downloaded and converted 3,350 FtP case documents (August 2017-June 2019) into text files.
• Employed non-negative matrix factorization (a machine learning topic analysis method) for data analysis.
• Analyzed cases at both uni- and multi-professional levels.

Main Results:

• Identified key topics including criminal offenses, dishonesty, drug offenses, English language proficiency, indemnity insurance, patient care issues (e.g., incompetence), and personal conduct (e.g., aggression, substance misuse).
• Patient care was the most frequent topic for dental, medical, and nursing professionals.
• Criminal offenses were the most frequent topic for pharmacy professionals.

Conclusions:

• While common themes exist across healthcare professions, specific priorities for fitness to practise concerns differ.
• Professional and educational organizations should tailor their strategies to address these profession-specific priorities.
• ML offers an efficient method for analyzing large datasets of FtP cases to inform regulatory and educational practices.