Abstract:

Objective The grey model was used to fit the incidence of tuberculosis and predict the epidemic trend of tuberculosis in Yining City, Xinjiang Uygur Autonomous Region, which provided a basis for local tuberculosis prevention and control.Methods Basing on incidence data of tuberculosis reported in Yining City, Xinjiang Uygur Autonomous Region in 2012—2018, grey models (“GM model”) were used to simulate incidences of tuberculosis. The incidences of tuberculosis in Yining City from 2012 to 2018 were 151.56/100000 (781/515299), 134.60/100000 (721/535668), 146.51/100000 (820/559691), and 152.00/100000 (893/587507), 142.64/100000 (782/548241), 177.85/100000 (992/558860), 196.81/100000 (1124/571122) respectively. The MATLAB software (produced by MathWorks, USA) was used to establish a GM model of first-order variable differential equation. The posterior difference ratio C and small error probability P were calculated to test the fitting of GM (1,1) model. The model fitting effect will be considered as acceptable if the C value is less than 0.5 and the P value is greater than 0.8, then the extrapolation prediction can be performed; otherwise, residual correction should be performed for the model.Results Our model was proved to be fitting well. The average residual of the model was 0.195, the average relative error was 4.553%, the small error probability P was 0.83, and the posterior difference ratio C was 0.43. The predicted incidences of tuberculosis in Yining City from 2019 to 2021 were 203.28/100000, 218.87/100000 and 235.66/100000 respectively.Conclusion The incidence of tuberculosis in Yining City from 2019 to 2021 may continue to increase with an annual increase rate of 7.67%. It is necessary to continue strengthening prevention and control of tuberculosis and reduce the harm of tuberculosis to the population.

Key words: Tuberculosis,pulmonary, Incidence, Epidemiologic studies, Models,statistical, Forecasting, Small-area analysis