The value of structured occupational safety data and cluster analysis: A case study from the Italian National Surveillance System

The collection and analysis of accident data are crucial steps towards the application of effective preventive measures in the occupational safety context. The analysis of work-related accidents/injuries is a focal starting point for improving the safety level in each sector. Data collection represents a critical activity in this field since it is not characterized by a standardized approach. Quantitative models allowing the extraction of root causes of an injury can be effectively applied without a vast computational effort only if structured safety data are available. The aim of this study is to demonstrate the benefits related to the adoption of a structured and standardized model for the collection and analysis of safety data. Thus, the application of quantitative methods (based on statistical models) to structured safety data about injuries is proposed in the present study. A case study on fatal injuries that occurred in the construction sector in Italy from 2002 to 2022 is discussed, based on data collected through a structured database called Infor.MO and managed by the Italian National Institute for Insurance against Accidents at Work (INAIL), Regions and Autonomous Provinces and Local Health & Safety Departments (LHSDs). A statistical analysis based on a multi-step process (including optimal scaling, principal component analysis, and cluster analysis) has been applied to point out the real accident scenarios and root causes characterizing the sector analyzed. Thanks to the structured set of data provided by Infor.MO, applying different statistical techniques has been supported, elaborating data and extracting meaningful information about the most significant fatal injury scenarios. The results obtained allowed us to point out four main clusters of fatal injury scenarios, highlighting common causes reliably. This application has pointed out operational and strategic benefits related to the availability of structured safety data to both highlight root causes of injuries and, in a more strategic way, to improve the effectiveness of prevention activities.