Exploration of Three Incidence Trend Prediction Models Based on the Number of Diagnosed Pneumoconiosis Cases in China From 2000 to 2019

Objective: To predict the future incidence trend of pneumoconiosis in China, and to evaluate three predictive models.

Methods: We selected pneumoconiosis cases (2000-2019) to fit Generalized Additive Model (GAM), Curve Fitting Method, and GM (1,1) Model, chosen average fitting relative error, relative error of prediction, and coefficient of determination to evaluate models.

Results: Chinese incidence trend of pneumoconiosis would decrease in the future. Predicted value of GAM (14,566) and Curve Fitting Method (15,781) in 2019 was close to the actual value (15,898). Relative error of prediction of GAM and Curve Fitting Method was -8.38% and -0.73%, respectively.

Conclusions: The government needs to strengthen prevention and control since pneumoconiosis cases might remain huge in the future. Besides, we advise that GAM and Curve Fitting Method can be used to predict Chinese incidence trend of pneumoconiosis.