The Prevalence, Risk Factors And Evaluation Of Intervention Effects Of Injuries Among Construction Workers

ObjectiveThe object of the study is to investigate non-fatal injuries, demographics, behavior trains and mental traits among construction workers. And based on these traits, we analyse potetional risk factors.According to the risk factors, workers'living and working environment, clinical medicine, preventive medicine, psychology and behavior science, we tend to build a set of health education method to change workers'behavior traits and mental traits to reduce injuries among workers.Method1. Epedemiological traits of occupational injuries and risk factorsWe utilized a purposive sampling method to select 1260 construction workers from three cities (Wuhan City of Hubei Province, Maanshan City of Anhui Province and Huzhou City of Zhejiang Province). We designed the survey questionnaire. The questionnaire collect detailed information about the following demographics and injury risk factors:personal demographics (including age, marital status, height, weight, school education years); behavioral risk factors (including cigarette smoking history and alcohol consumption); self-perceived sleep quality; self-perceived family economic status; psychological status and injury characteristics (external cause of injury, type of injury, major body part injured, month and place the injury occurred). We used frequncy and proportion to describe injuries with regard to demographics, self-perceived sleep quality, cigarette smoking status, alcohol consumption, and psychological status. We constructed logistic regression models to describe association between injury risk factors and injury status. We constructed exploratory factor analysis to describe association between factors. We used confirmatory factor analysis and structual equation models to construct models. We used artificial neural networks to predict injuries. The softwares used in the study included SPSS 13.0, SAS 9.0 and MATLAB2. Injury health education and evaluation of effectWe select 562 workers (the control group),478 workers (the intervention group) from 1260 workers. The intervention method was based on the theory of health belief model. The intervention had 5 sessions. We evaluate the effect of intervention after a year after intervention. We compared the injury characteristics before and after the intervention in control group and intervention. Generalized estimation equation was used for evaluating the intervention method and injuries.Result1. Injury prevalence and injury characteristicsThere were 189 workers with at least one work-related injury in all 1260 workers. The overall injury prevalence was 15.0 per 100 workers. The leading external causes of those injuries were collisions (27.32%). The leading type of injuries was superficial injuries (47.94%), followed by sprains and strainsand open wounds. Heads were the fewest body parts that occurred injuries. Body parts often injured were extremities and trunks, which accounted for 30% of all injuries, respectively. Most injuries occurred outdoors(58.76%),41.24% injuries occurred indoors. Most injuries occurred during the four-month period from July to October.2. Related factors of construction workers'injuriesLogistic regression analyses suggested that workers with the following risk factors were significantly more likely to suffer injuries than workers without those risk factors:high alcohol consumption, no injury prevention and safety education, and POMS depression. The results of structural equation model showed that obesity, social behaviors, social economical status, sleep quality, injury safety education and psychological status affected injuries directly. The accuracy of BP and Elman artificial neural networks was more than 60%.3. Results of intervention modelsThe study utilized psychological behavior and health belief intervention models. The injuriy prevalence of the control group was higher after a year. But the difference was not significant. The injuriy prevalence of the intervention group was lower after the intervention. But the difference was also not significant. The difference between injuriy prevalence of the control group and intervention group was significant after the intervention, resulting from the intervention. Using generalized estimation equation to evaluate the effect of the intervention, the result showed that the intervention method was significant.ResultOur study suggests that construction workers have a higher non-fatal injury prevalence than other workers. The high prevalence of work-related injuries among those construction workers may be the result of several risk factors, including serious alcohol consumption, not having injury prevention & safety education, and depressive symptoms. After the intervention, the injury prevalence of intervention group was significantly lower than that in control group. Except the intervention, this result also result from having injury prevention & safety education, and vigor symptoms.Innovation1. We systematically explored the prevalence of occupational injuries, related factors and intervention models. Based on the materials, we constructed logistic regression models, factor analysis, structural equation model to analyse related facotor, BP & Elman neural networks to predict injuies. And we used generalized estimation equation to evaluate the effect of the intervention.