基于Boruta算法和logistic回归模型的欧盟建筑业员工职业伤害影响因素分析

Influencing factors of occupational injury in construction workers of European Union based on Boruta algorithm and logistic regression

  • 摘要:
    背景 建筑业员工是职业伤害的高发群体。目前,国内外关于建筑业员工职业伤害影响因素的探讨多侧重于人口学以及行为特征等,对社会心理风险、数字技术使用以及员工健康状况则关注不足。
    目的 分析建筑业员工职业伤害的发生情况,探讨社会心理风险、数字技术使用、健康状况和工作场所预防措施等层面因素对职业伤害的影响,为预防措施的制定提供依据。
    方法 应用欧盟职业安全与健康管理局公开数据,样本包括2167名建筑业员工。以员工存在职业伤害为结局变量,通过卡方检验提取其中社会心理风险、对数字技术的使用、健康状况和工作场所预防措施等层面共25个变量,应用Boruta算法和多因素logistic回归分析模型相结合的方法,识别影响职业伤害的关键因素。
    结果 在调查的2167名建筑业员工中,存在职业伤害的有182人(占8.6%)。Boruta算法识别出8个重要特征变量,变量重要性由高到低依次是肌肉骨骼疾患、工种、抑郁与焦虑、受教育水平、使用电子智能产品、工作场所及时解决安全问题、全身疲劳、年龄;logistic回归分析显示:年龄、工种、受教育水平、全身疲劳、肌肉骨骼疾患以及工作场所及时解决安全问题共6个变量对职业伤害的影响有统计学意义(P<0.05)。
    结论 建筑业员工职业伤害的发生受多种因素影响,包括年龄、工种、受教育水平、全身疲劳、肌肉骨骼疾患以及工作场所及时解决安全问题。企业和员工应采取针对性的措施,减少职业伤害的发生。

     

    Abstract:
    Background Construction workers represent a high risk group for occupational injuries. Currently, domestic and international studies examining the factors affecting occupational injuries among construction workers focus on demographic and behavioural characteristics. However, there is limited attention to psychosocial, use of digital technology, and health status of workers.
    Objective To analyze the occurrence of occupational injuries among workers in the construction industry, explore impacts of psychosocial risk, use of digital technology, health status, and preventive measures at the workplace on occupational injuries, and provide a basis for the development of preventive measures.
    Methods Publicly available data from the European Union Occupational Safety and Health Administration were retrieved, comprising a sample of 2167 construction workers. The outcome indicator was the presence of occupational injuries among workers. A total of 25 variables in the dimensions of psychosocial risk, use of digital technology, health status, and preventive measures at the workplace were extracted after Chi-square test, and then a combination of Boruta's algorithm and logistic regression was applied to identify the key factors affecting occupational injuries.
    Results Among the 2167 construction workers surveyed, 182 (8.6%) reported experiencing occupational injuries. The Boruta algorithm identified eight characteristics which in descending order of importance were musculoskeletal disorders, job type, depression and anxiety, level of completed education, use of electronic smart products, timely solution of safety problems, overall fatigue, and age. The logistic regression results indicated that six variables had statistically significant effects on occupational injuries: age, job type, level of completed education, musculoskeletal disorders, overall fatigue, and timely solution of safety problems (P<0.05).
    Conclusion Occupational injuries in construction workers are influenced by a variety of factors, including age, job type, level of completed education, musculoskeletal disorders, overall fatigue, and timely solution of safety problems. Companies and workers should take targeted measures to reduce the incidence of occupational injuries.

     

/

返回文章
返回