Time series analysis of pulmonary tuberculosis in the elderly during 2006−2022 and short-term prediction in Henan province

  Objective  To understand the incidence of pulmonary tuberculosis (TB) from 2006 to 2022 and predict its incidence trend from 2023 to 2025 in the elderly in Henan province.

  Methods  The monthly pulmonary TB registration rates in the elderly in Henan from 2006 to 2022 were collected from “Tuberculosis Management Information System” to establish a database. The “expert modeler” module of software SPSS 27.0 was used to screen the optimal model. The monthly registration rate of pulmonary TB in the elderly from 2006 to 2019 was used as a training set, while that from 2020 to 2022 was used as a validation set. Then the registration rates of pulmonary TB in the elderly from 2023 to 2025 was predicted by the optimal model.

  Results  The average annual registration rate of pulmonary TB in the elderly in Henan from 2006 to 2022 was 147.06/100000, showing a downward trend. The monthly registration rates had a long-term fluctuating downward trend and seasonal characteristics. The best fitting model was ARIMA (1,0,0) (1,0,1)₁₂ model. The stationary R² and R² were both 0.857, close to 1, and the Ljung-Box Q (18) value was 12.156. The annual incidence of pulmonary TB in the elderly was high during March to June and in November. The predicted incidence of pulmonary TB in the elderly shows a slight upward trend from 2023 to 2025.

  Conclusion  Applying the ARIMA model, the monthly registration rate of pulmonary TB in the elderly is predicted to show a small increase from 2023 to 2025. It is necessary to establish a prevention and control mechanism for pulmonary TB in the elderly and strengthen the early detection and management of pulmonary TB in this population.