Explore the change trend fitting prediction model on new cases of occupational diseases in Guangdong Province / 中国职业医学

ABSTRACT

OBJECTIVE: To screen the optimal fitting model for the change trend of the number of new cases of occupational diseases in Guangdong Province by using linear and nonlinear regression models. Method The number of new cases of occupational diseases in Guangdong Province from 2003 to 2017 was used as the dependent variable(■) and the year(time) as the independent variable(x).Eleven mathematical models including linear regression, cubic function, quadratic function, composite function, growth function, exponential function, logistic function, power function, logarithmic function, S-type function and inverse function were used to fit the data, and the best-fit model was selected to describe and verify the change of new occupational diseases. RESULTS: Among the 11 mathematical models, the determination coefficient of fit results of cubic curve regression model was the highest(0.94, P<0.01), and the fit effect was the best. The fitting curve was ■. The cubic curve regression model was used to fit the number of new cases of occupational diseases in Guangdong Province from 2003 to 2019. The results showed that the measured value of new cases in all those years, except 2011, was within 95% confidence interval of the fitting value. The median(25 th, 75 th percentile) of absolute relative deviation between the fitting value and the actual value was 8.9%(4.3%, 14.7%). CONCLUSION: The regression model based on cubic curve can better fit the incidence of occupational diseases and can be used to describe the occurence of occupational diseases.