In occupational safety research, narrative text analysis has been combined with coded surveillance data to improve identification and understanding of injuries and their circumstances. Injury data give information about incidence and the direct cause of an injury, while near-miss data enable the identification of various hazards within an organization or industry. Further, near-miss data provide an opportunity for surveillance and risk reduction. The National Firefighter Near-Miss Reporting System (NFFNMRS) was a voluntary reporting system that collected narrative text data on near-miss and injurious events within the fire and emergency services industry.
The application of autocoding techniques using Bayesian models has been used to code injury narratives with up to 90% accuracy, thereby reducing the amount of human effort required to manually code large datasets. Autocoding techniques have not been applied to the NFFNMRS narrative data. Therefore, we applied machine learning algorithms and were successful in assigning mechanism of injury codes to near-miss and injury narratives from the NFFNMRS, reaching a sensitivity of 74%. Additionally, the autocoding was successful at assigning injury outcome (yes/no) to narratives, correctly assigning injury outcome to 92% of narratives. This created two new quantitative variables within the data system: mechanism of injury and injury outcome.
For deeper information, please consult the final report to NIOSH, the Accident Analysis and Prevention paper, and two other publications about the quality of data in the NFFNMRS and what it can achieve for injury risk assessment in firefighters, EMTs, and paramedics.