ABSTRACT
The addition of narrative text infonnation in electronic format to injury databases can be a useful adjunct to epidemiologic analyses and can suggest injury prevention strategies. Narrative fields provide many advantages if maintained in electronic format. Their ultimate use is to expand on coded injury data (e.g. exposure to noise) to better understand specific circumstances of an injury (Sorock et al., 1997). Grouping the data to satisfy investigative needs such as injury cause scenarios is an essential pari of the analytic process. Researchers at the Liberty Mutual Research Institute and Purdue University have been involved for over a decade in developing methods for semi-automatic classification of injury narratives into classifications which describe and group the circumstances of injuries. These "injury narratives" can be used to understand both non-injury and injury-producing incidents as well as injuries sustained by people involved in these events. In this paper we summarize some of the major findings of this research.
