Association Of Air Pollution With Daily Outpatient Visits To Hospital In Tianjin

In order to explore the effects of the air pollution on daily outpatient visits of hospital, the relation analysis and the regression analysis is used to analyze the relationship between the SO₂, NO_x, PM10 concentrations and the daily outpatient visits of the department of respiratory disease, internal medicine, emergency and otorhynolaryngology in Tianjin Fourth Central Hospital in 2010.Since there is Multicollinearity between independent variables, in addition to use stepwise regression, we should also take consider of principal component regression, ridge regression and partial least squares regression analysis. Moreover, we use the residual analysis and multicollinearity analysis to test the applicability of these models. We use the prediction residual sum of squares (PRESS) minimum standards to compares the accuracy of the model obtained. The results show that the precision of stepwise regression and principal component regression model is close, while the partial least squares regression and ridge regression models are significantly better than the other.The best fit models show that the regression coefficient of the three pollutants is all positive. However, the number varies widely. The regression coefficients of NO_x were relatively the largest where as the regression coefficients of PM10 is too small to be ignored. The results show that the integrated impact of NO_x and SO₂, especially the pollution of NO_x, is the statistically significant influential factor for the respiratory and internal medicine disease. NO_x is the main pathogenic factor of internal emergency disease as well as otorhynolaryngological disease.There is two main creative points. First, there is no research concerning on the association of air pollution with daily outpatient visits to hospital in Tianjin. Second, there is no quantitative study concerning on the relationship between the otorhynolaryngology disease and air pollution.