Causal factors and risk assessment of fall accidents in the U.S. construction industry: A comprehensive data analysis (2000–2020). (February 2022)

• Record Type: Journal Article
• Authors: Halabi, Yahia; Xu, Hu; Long, Danbing; Chen, Yuhang; Yu, Zhixiang; Alhaek, Fares; Alhaddad, Wael
• Is Part Of: Safety science. Volume 146(2022)
• Publication Date: 2022-02
• Subjects: Fall accidents -- Accident prevention -- Trends analysis -- Logistic regression -- Project management -- Safety monitoring -- Risk assessment; Industrial accidents -- Periodicals; Accident Prevention -- Periodicals; Safety -- Periodicals; Travail -- Accidents -- Périodiques; 363.11
• DOI: 10.1016/j.ssci.2021.105537
• Languages: English
• ISSNs: 0925-7535

Abstract

Highlights: Fall accident trends in the U.S. construction industry can be identified in this study, relying on reliable data for statistical analysis. A profound understanding of the fall accident factors provides effective prevention strategies. The analyzed data can help the managers to prioritize tasks for workers on-site to improve safety. This study employed the prominent accident factors to develop a prediction model which can diagnose the fall risks fatal and nonfatal likelihood. Abstract: This study delves into investigating the leading factors of occurring 23, 057 fall accidents in the United States construction industry over 20 years (1/2000–8/2020) recorded in the Occupational Safety and Health Administration (OSHA) database. Additionally, the contributions are elicited in terms of diverse dimensions of fall accident, such as project type, construction end-use, work activity, worker's occupation and age, fall location and height, accident time, injury degree, and fall protection. The data is analysed using frequency analysis to obtain the trends of fall accidents, correlation analysis between the accident factors and the injury degree, and logistic regression analysis to establish a prediction model that can diagnose fatal and nonfatal accidents. The results emphasized that the proportion of fall accidents increased substantially, and there was egregious evidence that the usage of fall protection has no considerable improvement. Besides, most of the fall accidents were (1) from heights<9.15 m, (2) among the roofers, (3) occurring on new commercial buildings and residential projects with low cost, (4) during the time intervals 10:00–12:00 and 13:00–15:00, (5) among older workers which alert that the experience might not be enough to diminish the accident. The correlation analysis revealed the fall factors that were significantly associated with the injury degree. Subsequently, a logistic regression model was done to predict the injury outcome (fatal/nonfatal). It was found that the prediction model could correctly diagnose the injury degree outcome by 77.7% depending on the selected predictors of the fall accident. Furthermore, the odds of reporting fatal or nonfatal accidents from the prominent factors of fall were calculated, enhancing the risk assessment to avoid the implications of falls. This study might encourage the safety managers to apply proactive and preparedness procedures for reducing fall accidents and prioritize risks according to the likelihood of fall risk and injury characteristics by applying appropriate safety regulations.