Abstract Spearman's rank correlation, Cramer's V and Phi coefficients were applied to the analysis of 784 work-related single fatalities that occurred in 1999 and 2000. The Phi coefficients were used to screen for significant accident scenarios in terms of industry, source of injury and accident type, and high-risk groups in terms of company size, worker's experience and industry group combinations. Significant accident scenarios identified included: fatalities from falls caused by structure and construction facility in the construction industry; caught in between and clamped fatalities caused by loading and unloading machinery, and power machinery in the manufacturing industry; explosion and contacting hazardous materials and extreme temperature fatalities caused by materials and supplies in the manufacturing industry; struck by and against fatalities caused by loading and unloading machinery in the transport, storage and communication industry; and drowning fatalities caused by environment in the farming and fishing industry. High-risk groups identified were construction workers with less than 1 year of experience who were employed by small companies with less than 30 workers; and manufacturing workers with 1–15 years of experience who were employed by large companies with more than 30 workers. Relevance to industry Significant accident scenarios and high-risk groups were identified based on the analysis of 784 work-related single fatalities. These results provided a direction for in-depth scenario analysis and for more effective inspection strategies which may reduce both accident frequency and severity.