季节趋势模型在犬伤门诊就诊人次数预测中的探究  

An investigation of seasonal trend model in the prediction of the number of outpatient visits in dog injury clinics

作　　者：蒋莹[1] 徐哲[1] 程庆林[2] 黄仁杰[2] 娄静 周卫群[1] JIANG Ying;XU Zhe;CHENG Qinglin;HUANG Renjie;LOU Jing;ZHOU Weiqun(Jiande Center for Disease Control and Prevention,Jiande,Zhejiang 311600,China;不详)

机构地区：[1]建德市疾病预防控制中心,浙江建德311600 [2]杭州市疾病预防控制中心

出　　处：《医学动物防制》2024年第6期559-562,共4页Journal of Medical Pest Control

基　　金：浙江省基础公益研究计划项目(LGF21H260007);杭州市农业与社会发展科研引导项目(20220919Y059)。

摘　　要：目的探讨季节趋势模型在犬伤门诊人次数预测中的应用,为制定防控措施和策略提供科学依据。方法应用SPSS 20.0和Excel 2007对建德市2017—2022年各月份的犬伤门诊人次数建立预测模型(X_(t)=Y_(c)×S_(t)),以月份序号与犬伤门诊人次数建立直线回归方程求出趋势值Y_(c),根据2017—2022年各月份的季节指数S,计算1~12月的平均调整季节指数S_(t),将2017—2022年各月份的犬伤门诊人次数拟合值与实际值进行拟合评价,并对2023年犬伤门诊人次数进行预测。结果建德市每月均有狂犬病暴露患者就诊,总体就诊高峰为5~8月,低谷在11月至次年3月,存在季节波动。直线回归方程Y_(c)=569.41-1.86t,方差分析结果显示,回归方程检验差异有统计学意义(F=10.007,P<0.05)。季节趋势模型拟合的平均相对误差为7.55%,模型拟合较好,预测2023年犬伤门诊人次数为5080人,月预测值分别为317、362、374、446、505、509、524、505、437、432、359、310。结论季节趋势模型可用于犬伤门诊人次数的短期预测,该方法对狂犬病的预防有重要意义,其原理简单,方法简便易于掌握,适合基层应用。Objective To explore the application of seasonal trend model in the prediction of the number of outpatient visits in dog injury clinics,and to provide a scientific basis for the development of prevention and control measures and strategies.Methods A prediction model(X_(t)=Y_(c)×S_(t))was established for the number of outpatient visits in dog injury clinics in each month from 2017 to 2022 in Jiande City using SPSS 20.0 and Excel 2007.A linear regression equation was established to determine the trend value Y_(c) using the month serial number and the number of outpatient visits in dog injury clinics.An average adjusted seasonal index S_(t) was calculated from January to December according to the seasonal index S of every month from 2017 to 2022.The fitted values and actual values for the number of outpatient visits in dog injury clinics for each month from 2017 to 2022 were assessed by fit evaluation,and the number of outpatient visits in dog injury clinics were projected for 2023.Results There were patients exposed to rabies visiting a doctor in Jiande City every month,with an overall peak from May to August and a trough from November to March,indicating seasonal fluctuations.The linear regression equation was Y_(c)=569.41-1.86t,and the results of the analysis of variance showed that the difference in the regression equation test was statistically significant(F=10.007,P<0.05).The seasonal trend model had a relatively good fit,with an average relative error of 7.55%.In 2023,the number of outpatient visits in dog injury clinics were predicted to be 5080,with monthly projected values of 317,362,374,446,505,509,524,505,437,432,359,and 310,respectively.Conclusion The seasonal trend model is used for short-term prediction of the number of outpatient visits in dog injury clinics and is important to prevent rabies.The principle is simple,and the method is easy to grasp,which is suitable for grassroots application.

关 键 词：季节趋势模型 犬伤门诊 狂犬病 季节指数 短期预测 门诊人次 预测 探究 

分 类 号：R183[医药卫生—流行病学] R512.99[医药卫生—公共卫生与预防医学]