Classification of contributory factors in free-text NHS patient safety incident reports

The extraction of contributory factor information from patient safety incident reports is needed for system-level feedback into learning healthcare organisation. Towards this end, this project seeks to apply natural language processing (NLP) and supervised machine learning (ML) approaches to the automated classification of patient safety incident reports based on free-text descriptions. The dataset to be used is derived from England and Wales National Reporting and Learning System data that has been clinician-coded. The endpoint would be the production of a classifier that could assign contributory factors to previously unseen patient incident free-text.