摘要
目的利用指数平滑法拟合并预测手术部位感染发生趋势,为医院手术部位感染的防控工作提供科学依据。方法回顾性分析2015-01-30-2020-06-30赤峰学院附属医院每月的手术部位感染数据,使用SPSS 22.0建立季节性指数平滑模型,并预测2020年下半年医院手术部位感染率。结果拟合最佳的季节性指数平滑模型为简单季节模型,平稳R^(2)=0.726,平均绝对百分比误差(MAPE)=21.281,模型残差为白噪声序列(Ljung-Box Q=22.354,P=0.132),拟合值与真实值曲线较为接近,实际值均落在预测值95%CI范围内,提示模型具有较好的拟合能力。2020年下半年各月手术部位感染率预测值分别为1.05%,95%CI(0.64%~1.46%);1.06%,95%CI(0.64%~1.48%);0.77%,95%CI(0.33%~1.21%);0.75%,95%CI(0.29%~1.20%);0.52%,95%CI(0.05%~0.99%);0.53%,95%CI(0.05%~1.02%)。结论简单季节模型能够很好地预测手术部位感染发生趋势,可为医院手术部位感染相关防控措施的制定提供参考。
Objective To fit and predict the trend of infection at the surgical site using the exponential smoothing method,so as to provide scientific basis for prevention and control of surgical site infection.Methods By means of SPSS 20.0,the seasonal exponential smoothing method was established for the monthly incidence of surgical site infection from Jan 12015 to Jun 302020.Results The simple seasonal model was finally identified as the optimal model.Stationary R^(2)=0.726,MAPE=21.281,and the residuals of the model were white noise(Ljung-Box Q=22.354,P=0.132).The fitted value is closer to the true value curve,and all the actual values were within the 95%confidence interval of the fitted value,which suggests that the model has a better fitting ability.The predicted incidence rates of surgical site infection were 1.05%,95%CI(0.64%-1.46%);1.06%,95%CI(0.64%-1.48%);0.77%,95%CI(0.33%-1.21%);0.75%,95%CI(0.29%-1.20%);0.52%,95%CI(0.05%-0.99%);0.53%,95%CI(0.05%-1.02%),respectively.Conclusion The simple seasonal model can well fit the trend of infection at the surgical site,which can provide reference for the establishment of related prevention and control measures for hospital surgical site of infection.
作者
唐琳
TANG Lin(Affiliated Hospital of Chifeng University,Chifeng 024005,China)
出处
《社区医学杂志》
CAS
2021年第22期1370-1374,共5页
Journal Of Community Medicine
关键词
手术部位感染
指数平滑法
医院感染
预测
surgical site infection
exponential smoothing method
nosocomial infection
forecast
作者简介
通信作者:唐琳,女,内蒙古赤峰人,硕士,主治医师,主要从事疾病预防与控制的研究工作。E-mail:tanglin0416@163.com
