Abstract: Objective To establish the tuberculosis epidemic situation prediction model.Methods Aiming at the limitations of Aruma tuberculosis model,the tuberculosis epidemiology in china had been translated into a model including series of equations. The evolution of tuberculosis epidemic situation for 1979-2000 in Beijing had been simulated with the new established model.And the simulated data was compared with the actual data.Goodness-of-model fitting has been evaluated.Results The simulated data numbers of registered active tuberculosis patients number, registered sputum positive tuberculosis patients, tuberculous infection rate, and the tuberculosis prevalence from national tuberculosis sampling epidemiological survey were fitted with the acual datas. The Results of sensitivity analysis of the key factors in model indicated that only the proportion of tuberculosis patients managed by the Tuberculosis Control Stations elevated, then the DOTS strategy could affect the epidemic situation significantly.Conclusion The tuberculosis epidemic situation prediction model could reflect the actual tuberculosis evolution exactly, and was fit for tuberculosis control measures evaluation and epidemic situation prediction.

Key words: Pulmonary tuberculosis/epidemiology, Mathematical model, Prediction