MUSCULOSKELETAL WORK INJURY--EPIDEMIOLOGY AND DISABILITY

点击翻译按钮获取中文摘要

This five year project will investigate compensable (occupational) musculoskeletal work injuries and related neurologic injury, grouped by the part of body injured. The project will study the following injuries: contusions, crush injuries, bruises; sprains, strains; dislocations; fractures; repetitive motion injuries in the form of inflammation or irritation of the joints, muscles or tendons; and carpal tunnel syndrome. The injury areas include: back; upper extremity -- shoulder, upper arm/forearm, elbow, wrist, hand; lower extremity -- hip, thigh/lower extremity, knee, ankle and foot. The aims of the project are: 1) to describe the epidemiology of these conditions; 2) to characterize the resulting disability measured by the cumulative duration of missed workdays; and, 3) to perform hypothesis testing to determine whether gender, age, source of injury, type of accident, industry, occupation and employment size of the workplace are risk factors for injury occurrence and to estimate their contribution, if any, to cumulative missed worktime. The long-term project goals are to use the results to rethink and fine-tune current prevention and rehabilitation strategies from the perspective of the impact of occupational injuries on the entire workforce of a major industrialized state. The project uses data on 42,773 musculoskeletal injuries contained in the 1986 Michigan Database. The Michigan Database links missed workday information from the workers' compensation system with an average follow-up of more than three and a half years from the injury date, with injury characteristic information coded as part of the federal Bureau of Labor Statistics Supplementary Data System. The Michigan Database appears to be unique in providing longitudinal data for the entire workforce of a major industrialized state. Preliminary analyses indicate substantial age and gender effects in the disability produced by back sprains/strains and repetitive motion injuries, respectively. The following experimental design and methods will be used. To achieve the first project aim, incidence rates and rate ratios, their standard errors and 95% confidence intervals, cross-classified by the various covariables, will be calculated. For the second, a set of tables giving the quantile distributions for missed worktime for finely detailed cross-classifications of the various covariables will be created. To achieve the third aim, the testing of hypotheses and quantitation of effects, if any, of the predictor variables, multiple regression analyses will be performed. First-order interactions will be included in the regression models and, if confirmed, stratified main effects analyses will also be reported. Because of the finely detailed taxonomy of the covariables, various statistical techniques will be used to correct for potential multicollinearity. In addition, special attention will be paid to biologic plausibility and causality issues in aggregating the categorical variables used to represent the subgroups of a covariable.