[Impact of official front-end software for intelligent surveillance and early warning of infectious diseases on reporting of intestinal infectious diseases in China: an empirical analysis based on prediction models].

Objective: To evaluate the impact of the official front-end software for intelligent surveillance and early warning of infectious diseases on the reporting of other infectious diarrheal disease, hand foot and mouth disease (HFMD), and bacillary dysentery in 113 medical institutions at or above grade Ⅱ in Beijing, China. Methods: Based on monthly incidence data from 2012 to 2023, we constructed multiple prediction models, including seasonal autoregressive integrated moving average (SARIMA) model, error-trend-seasonality model, neural network auto regressive (NNAR) model, Prophet model, Bayesian structural time series model and hybrid model. The data in 2024 were used as the test set for the model evaluation, and the optimal prediction models were used to predict the case count from January to August 2025 without the intervention of front-end software use. Changes in reporting case count were analyzed by comparing actual reporting with prediction from January to August 2025 and actual reporting during the same period in 2024 by using paired t-test or Wilcoxon signed-rank test. Results: The optimal prediction models for other infectious diarrheal diseases and HFMD were SARIMA (1,0,1)(1,1,1)12 and SARIMA (3,0,0)(0,1,1)12 respectively, while the optimal prediction model for bacillary dysentery was NNAR (3,1,2)12. After the use of front-end software, the actual reported case count of other infectious diarrheal diseases increased by 1 625.75 cases compared with the prediction from January to August 2025, an increase of 168.71% (P<0.001) and by 1 500.25 cases compared with the actual case count during the same period in 2024, an increase of 152.61% (P<0.001). The actual reported case count of HFMD increase by 103.38 cases compared with prediction from January to August 2025, an increase of 75.80% (P=0.008), but showed no significant difference compared with the same period in 2024 (P=0.461). No significant changes were observed in the reporting bacillary dysentery before and after the front-end software use in the two comparisons (P=0.895 and P=0.239). Conclusion: The application of the official front-end software significantly improved the reporting sensitivity for other infectious diarrheal diseases and HFMD, but no significant effect was observed in the reporting of bacillary dysentery.