Work-related overexertion injuries in cleaning occupations: An exploration of the factors to predict the days of absence by means of machine learning methodologies.

The special characteristics of the cleaning industry have an important impact on the health and safety of its workforce. Making use of data from more than 79,000 occupational accidents, the aim of the present research is to use machine learning techniques to develop a model to predict incapacity for work (expressed in days of absence) due to work-related overexertion injuries among service sector cleaners in Spain. The severity of accidents caused by overexertion depends on several factors that can be classified into the following categories: injury typology, individual factors, employment conditions, accident circumstances and health and safety management and standards in the company.

Computer vision syndrome in healthcare workers using video display terminals: an exploration of the risk factors.

AIMS: To estimate the prevalence of computer vision syndrome (CVS) in healthcare workers and its relationship to video display terminal (VDT) exposure, sociodemographic, optical correction and work characteristics, and to analyse whether there are differences among occupational groups. DESIGN: Cross-sectional study. METHODS: A sample of 1179 physicians and surgeons, nurses, and nursing assistants from two hospitals in Spain between January 2017 and February 2018 were invited to participate in this study. Of these, 622 workers from both hospitals were finally included. CVS was measured using a questionnaire, the CVS-Q© . Logistic regression was used to identify the factors associated with CVS. All the results were stratified by occupational group. RESULTS: The prevalence of CVS was 56.75% with nurses being the most affected occupational group (61.75%). It was associated significantly with female sex (aOR = 2.57; 95% CI 1.36-4.88) and morning shifts plus on-call (aOR = 2.33; 95% CI 1.11-4.88) in the physicians and surgeons group. Among the nurses, it was associated with female sex (aOR = 2.35; 95% CI 1.03-5.37), seniority between 10 and 20 years (aOR = 2.17; 95% CI 1.03-4.59), VDT exposure at work of 2-4 h/day (aOR = 6.14; 95% CI 1.08-35.02), VDT exposure at work >4 h/day (aOR = 7.14; 95% CI 1.29-39.62) and self-perception that using the software application was not easy (aOR = 2.49; 95% CI 1.23-5.01). CONCLUSIONS: A high prevalence of CVS among healthcare workers was observed. The risk factors that increased the likelihood of suffering from this syndrome depended on the occupation. IMPACT: The findings may be used as a reference for occupational health services to implement specific preventive measures to reduce CVS for each occupational group. Such measures should consider both individual factors and the working conditions.

Prediction of Computer Vision Syndrome in Health Personnel by Means of Genetic Algorithms and Binary Regression Trees.

One of the major consequences of the digital revolution has been the increase in the use of electronic devices in health services. Despite their remarkable advantages, though, the use of computers and other visual display terminals for a prolonged time may have negative effects on vision, leading to a greater risk of Computer Vision Syndrome (CVS) among their users. In this study, the importance of ocular and visual symptoms related to CVS was evaluated, and the factors associated with CVS were studied, with the help of an algorithm based on regression trees and genetic algorithms. The performance of this proposed model was also tested to check its ability to predict how prone a worker is to suffering from CVS. The findings of the present research confirm a high prevalence of CVS in healthcare workers, and associate CVS with a longer duration of occupation and higher daily computer usage.