Developing a Geospatial Framework for Severe Occupational Injuries Using Moran’s I and Getis-Ord Gi* Statistics for Southeastern United States

Abstract

Occupational safety and health (OSH) related agencies have a plethora of injury data available; however, modern data visualization and user-friendly dissemination tools are lacking in this field. This work has created a geospatial platform using geographic information systems (GIS) combined with big data analytics for the southeastern region of United States. Severe injury reports were collected and visualized through state-of-the-art spatial statistics including spatial clustering, heat maps, and hotspots to identify the most vulnerable regions due to various injury types from a safety vantage point. A special focus was also given to the construction industry, considering the hazardous nature of the industry. Statistically significant spatial clustering was observed within the study region, with Moran’s I index of 0.49. Hotspots for severe injuries were also identified with approximately 99% confidence level using Getis Ord Gi* statistics. Results indicated statistically significant high-risk zones particularly around city areas with spatial injury rates (SIR) up to approximately 2.84 per county. Analysis also showed increased number of severe injuries during summer months, with approximately 1,000 injuries during the month of June and July. The construction industry accounted for 20% of all injuries, with “caught-in/between” being the highest amongst the four primary causes of severe injuries, commonly known as the “fatal four.” Finally, a web-based application was created to disseminate the results. Big data mining coupled with geospatial technology in OSH management can offer decision makers up-to-date and highly geospatial information for severe injuries that can be integrated in developing a comprehensive occupational safety surveillance plan.