Comparing the causes of work-related injuries in Australia, New Zealand and the United States using narrative information.
Williamson-A; Feyer-A-M; Stout-NA; Driscoll-T
NOIRS 2000--Abstracts of the National Occupational Injury Research Symposium 2000, Pittsburgh, PA, October 17-19, 2000. Pittsburgh, PA: National Institute for Occupational Safety and Health, 2000 Oct; :79-80
Narrative or text-file information is often collected to supplement injury data in large-scale datasets. Often, however, this style of information is not analysed because of uncertainty about the quality of the analysis. The opportunity arose in this project to investigate systematically the effects of coding of the circumstances of fatal occupational injury using narrative analysis compared to analysis using standard methods. This project was part of the three country collaborative study of occupational fatalities, which compared population data from New Zealand, Australia, and the USA. In these datasets, both Australia and New Zealand used the same standard code for mechanism of injury, but the USA did not, yet all countries included narrative information on the circumstances of the injury in their datasets. It was possible, therefore, to develop a narrative-based text search using the existing mechanism coding for Australia and New Zealand which was maximally accurate in reflecting the already-coded mechanism information. It was then possible to apply the narrative search to the US data as well, in order to compare the circumstances of injury in each of the three countries. The results showed that narrative-based coding produced the same patterns as the standard coding. Hit by moving objects and falls were the most common mechanisms of injury for each country. Some types of mechanisms, however, could be represented more accurately than others. The main errors in the narrative analysis were lack of sensitivity in picking up cases rather than poor specificity of coding.
Accident-statistics; Accidents; Accident-prevention; Injuries; Traumatic-injuries; Injury-prevention; Surveillance-programs; Statistical-analysis; Epidemiology
Conference/Symposia Proceedings; Abstract
NOIRS 2000 Abstracts of the National Occupational Injury Research Symposium 2000, Pittsburgh, PA., October 17-19, 2000