Based on 39305 cases of outpatients from 2001 to 2002 in two big hospitals of Nanning, it is concluded by using wavelet analysis that there is a 5 -7 day period in the number of patients. The relationship among respiratory tract disease, meteorological factors and pollutant concentrations is analyzed by using the data on the outpatients and the corresponding meteorological conditions and air pollutants. The results show that the number of respiratory tract disease patients is closely correlative with air temperature, air pressure and humidity, as well as the concentrations of such air pollutants as SO2, NO2, and PM10. Meanwhile, the optimal subset method was used to establish the regressive equations for forecasting respiratory tract diseases in the next 1 to 2 days and the next week for different seasons. The regressive equations were tested based on the bronchitis cases of outpatients in January 2004 and the respiratory tract disease cases of outpatients in April 2004, and the results show that the regressive equations are applicable.