关键词: 猩红热, 流行病学特征, 季节性自回归综合移动平均模型, 趋势预测

Abstract: Objective To understand the epidemiological characteristics and changing trend of scarlet fever in Shapingba District of Chongqing,provide a basis for developing prevention and control measures. Methods The descriptive epidemiological method was used to analyze the epidemic data of scarlet fever among children under 15 years old in Shapingba District of Chongqing City from 2014 to 2023. The seasonal autoregressive integrated moving average model（SARIMA） was constructed for predicting the epidemic trend of scarlet fever. Results A total of 725 cases of scarlet fever in children under 15 years old were reported in Shapingba District of Chongqing City from 2014 to 2023,with the average annual incidence rate of 54.17/100 000,and the incidence rate in 2019 was the highest（119.79/100 000）. Since 2014,the incidence rate showed an upward trend（χ²_trend=43.448,P<0.01）. Odd years had higher incidence intensity than even years. The epidemic showed a bimodal distribution,which were April to July and October to January of the following year respectively. During the peak period,there were 620 cases,accounting for 85.38% of the total number of cases. The number of cases in the eastern urban area（389 cases,53.66%） was significantly higher than that in the western new urban area,with a statistically significant difference（χ²=4.053,P<0.01）. The incidence rate of boys（59.42/100 000） was higher than that of girls（48.74/100 000）,with a statistically significant difference（χ²=7.036,P<0.05）. The cases were mainly children aged 3-9 years old（627 cases,86.48%）. Most of cases were students and kindergarten children（683 cases,94.21%）. The model of SARIMA（2,1,3）（1,0,2）₁₂ was established. The coefficient of determination（R²）,the normalized Bayesian information criterion（BIC） and the root mean squared error（RMSE） were used to evaluate the goodness-of-fit of the fitted model. The R²,BIC and RMSE of predictive value was 0.791,1.731 and 1.634,showing good fit with the actual value. The prediction results showed that there would be two peaks of incidence from July 2024 to June 2025,with peaks of 13.44/100 000 and 13.90/100 000 respectively,which were higher than the same period level. Conclusion Since 2014,the scarlet fever epidemic shows a significant upward trend,with odd years being higher than even years. The students aged 3-9 years old（preschool children） are at high risk of scarlet fever,and the seasons of late spring and early summer and late autumn and early winter are the peak periods of the epidemic. The model predicts that the epidemic will increase in 2025. It is necessary to continue to strengthen the monitoring and prevention of scarlet fever based on the characteristics of the epidemic,and take targeted prevention and control measures.

Key words: Scarlet fever, Epidemiological characteristics, Seasonal autoregressive integrated moving average model, Trend prediction