RESUMO
The UK's work-related diseases and occupational injury surveillance system consists of Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 2013 (RIDDOR), Labour Force Survey (LFS), The Health and Occupation Research network in General Practice (THOR-GP), and Reporting to The Health and Occupation Research network by specialist physicians. This article briefly described the scope, content, and methods of each surveillance programme in the UK work-related diseases and occupational injury surveillance system, and summarized their advantages and disadvantages. Among them, employers are required to report to relevant law enforcement authorities by RIDDOR, data are highly accessible but with a concern of serious underreport, and it is the only data source of fatal occupational injuries; LFS, a representative national household sample survey, covering occupational injuries and work-related diseases, is the primary data source of non-fatal occupational injuries and work-related diseases such as stress, anxiety, and depression, but collects non-clinically proven data based on self-perception; general practitioners report clinically confirmed work-related diseases, which is more scientific in attribution and is a good secondary source of work-related diseases; specialist physicians report clinically confirmed cases of higher severity, which is the primary source of data on conditions such as asthma and dermatitis, but may underestimate morbidity. Each surveillance programme of the system has its own characteristics, intersects, and complements each other, which can provide reference for the construction of occupational injury surveillance system in China.
RESUMO
Background The severity of occupational injury in countries such as the United Kingdom, the United States, and Germany is usually analyzed using lost workdays, but in existing occupational injury surveillance research in China, the application of this index is rare. Objective To evaluate the application value of lost workdays in non-fatal occupational injury surveillance, and provide a reference for the construction of occupational injury surveillance index system. Methods The public data of European Statistics on Accidents at Work (ESAW) from 2010 to 2019 on non-fatal injury accidents in 27 member states of the European Union were used. Non-fatal occupational injury is defined as an injury event during occupational activities or at work resulting a victim's absence from work for ≥4 d. According to the European Statistics on Accidents at Work-Summary methodology, the lost workdays were divided into 8 categories (4-6 d, 7-13 d, 14-20 d, 21-30 d, 31-91 d, 92-182 d, 183 d and above, and unknown). Annual percentage change (APC) and the average annual percentage change (AAPC) were used to evaluate the overall trend changes in the incidence rate of non-fatal occupational injury accidents in different lost workdays from 2010 to 2019, and the non-fatal occupational injury accidents in key industries. The characteristics of the occurrence of non-fatal occupational injuries were analyzed in conjunction with the changes in non-fatal occupational injuries in different lost workdays in the industry. Results From 2010 to 2019, the overall incidence of non-fatal occupational injury accidents in the European Union showed a downward trend, and the AAPC was −1.0% (P<0.05). The accident rates of lost workdays of 4-6 d and 92-182 d showed an upward trend, and the AAPC were 7.9% and 5.8% respectively (P<0.05). The average annual accident rates of non-fatal occupational injuries (≥4 d) in Categories C (manufacturing industry), E (water supply, sewage treatment, waste management and remediation), and F (construction industry) showed a linear downward trend, and the AAPC were −3.0%, −2.5%, and −1.5%, respectively (P<0.05). However, among them, the rate of non-fatal occupational injury accidents with 92-182 d of lost workdays in the manufacturing industry showed a significant upward trend, with an AAPC of 3.7% (P<0.001). Conclusion Using lost workdays combined with APC and AAPC by Join-point linear regression analysis can measure the severity and trend changes of non-fatal occupational injury accidents in different industries and different lost workdays. This indicator has an important practical significance in evaluating the effectiveness of occupational injury prevention and control strategies adopted by countries and enterprises.
RESUMO
Background Identification and analysis of influencing factors of occupational injury is an important research content of feature selection. In recent years, with the rise of machine learning algorithms, feature selection combined with Boosting algorithm provides a new analysis idea to construct occupational injury prediction models. Objective To evaluate applicability of Boosting algorithm-based model in predicting severity of miners' non-fatal occupational injuries, and provide a basis for rationally predicting the severity level of miners' non-fatal occupational injuries. Methods The publicly available data of the US Mine Safety and Health Administration (MSHA) from 2001 to 2021 on metal miners' non-fatal occupational injuries were used, and the outcome variables were lost working days < 105 d (minor injury) and ≥ 105 d (serious injury). Four different feature sets were screened out by four feature selection methods including least absolute shrinkage and selection operator (Lasso) regression, stepwise regression, single factor + Lasso regression, and single factor + stepwise regression. Logistic regression, gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost) were selected to construct prediction models by training with the four feature sets. A total of 12 prediction models of severity of miners' non-fatal occupational injuries were built and their area under the curve (AUC), sensitivity, specificity, and Youden index were calculated for model evaluation. Results According to the results of four feature selection methods, age, time of accident occurrence, total length of service, cause of injury, activities that triggered injury occurrence, body part of injury, nature of injury, and outcome of injury were identified as influencing factors of non-fatal occupational injury severity in miners. Feature set 4 was the optimal set screened out by single factor+stepwise regression and the GBDT model presented the best predictive performance in predicting the severity of non-fatal occupational injuries. The associated specificity, sensitivity, and Youden index were 0.7530, 0.9490, and 0.7020, respectively. The AUC values of logistic regression, GBDT, and XGBoost models trained by feature set 4 were 0.8526 (95%CI: 0.8387, 0.8750), 0.8640 (95%CI: 0.8474, 0.8806), and 0.8603 (95%CI: 0.8439, 0.8773), respectively, higher than the AUC values trained by feature set 2 [0.8487 (95%CI: 0.8203, 0.8669), 0.8110 (95%CI: 0.8012, 0.8344), and 0.8439 (95%CI: 0.8245, 0.8561), respectively] . The AUC values of GBDT and XGBoost models trained by feature set 4 were higher than that of logistic regression model. Conclusion The performance of the prediction models constructed by predictors screened out by two feature selection methods is better than those by single feature selection methods. At the same time, under the condition of optimal feature set, the performance of model prediction based on Boosting is better than that of traditional logistic regression model.