Abstract: Abstract: Objective   To establish a temporal prediction model for understanding the impact of COVID-19 on incidence of scarlet fever in Songjiang District of Shanghai. Methods   Scarlet fever cases in Shanghai area reported during 2011 and 2020 were retrieved from the data module of Infectious Disease Surveillance under the Chinese Disease Prevention and Control System. Seasonal autoregressive integrated moving average (SARIMA) was used to establish the forecasting model based on the number of monthly reported cases of scarlet fever between January 2011 and June 2019, and then the model was validated for the predicting efficacy in compliance with the real case numbers of scarlet fever from July to December of 2019. The optimal model was used to predict the occurrence of scarlet fever in 2020, and the impact of COVID-19 on the total number of scarlet fever in Songjiang District was analyzed. Results   In total, 4 438 cases of scarlet fever were reported in Songjiang District during 2011 and 2020. Averagely, 444 cases were annually reported, with an annual incidence rate of 23.27/100 000. The prevalence was represented in seasonal and bimodal distribution. The SARIMA (1,1,0)(2,1,1)12 model was eventually selected as the best-fit model out of the plausible models, with AIC, BIC and MAPE being 90.93, 105.93 and 7.68%. The residual test demonstrated a white noise sequence, and that the predicted value was basically in line with the actual value trend. A total of 94 cases of scarlet fever was identified in Songjiang District in 2020 due to the impact of COVID-19, which was 84.09% lower than expected value. Conclusion   The SARIMA model can fit the changing trend of scarlet fever cases in Songjiang District. And prevalence of scarlet fever tended to decrease sharply and be lower in Songjiang area in 2020 because of the impact from COVID-19. 

Key words: COVID-19, Scarlet fever, Seasonal autoregressive moving average model, Prediction, Songjiang District