基于时间序列模型的北京市院前急救出车车次预测分析  

Prediction and analysis of the number of pre-hospital emergency ambulance trips in Beijing based on the time series model

作　　者：邓贵芳 孙涛 耿聆[1] 巴衣尔策策克 陈辉[1] DENG Guifang;SUNTao;GENGLing;Bayierceceke;CHEN Hui(Beijing Emergency Medical Center,Beijing 100031,China)

机构地区：[1]北京急救中心,北京100031

出　　处：《中国急救复苏与灾害医学杂志》2024年第5期587-590,共4页China Journal of Emergency Resuscitation and Disaster Medicine

基　　金：首都卫生发展科研专研(编号:2024-2G-3031)。

摘　　要：目的研究分析北京市院前急救出车车次,预测未来时间内的出车车次,以期为北京市院前急救建设及发展提供参考。方法采用描述性统计方学法分析北京市2018年—2022年院前急救出车车次基本情况;使用SPSS 26.0统计软件建立时间序列模型,运用“专家建模器”自动选择最优模型,对北京市2023年院前急救出车车次进行预测。结果北京市2018年—2022年院前急救出车车次逐年上升,2022年出车车次是2018年出车车次近2倍,每年出车车次最高、最低月份分别为12月、2月;时间序列模型自动选择最优模型为“温特斯加型”,模型拟合度R方为0.896,平稳R方为0.377,杨-博克斯Q(18)统计量的显著性P值为0.642,数据拟合效果良好;预测值与实际值平均绝对百分比误差(MAPE)为6.85%,模型的预测能力“优良”;较好预测了2023年院前急救出车车次。结论北京市院前急救出车车次呈逐年上升趋势,院前急救公共卫生服务能力有效提升;时间序列模型较好地拟合北京市院前急救出车车次变化趋势并进行预测,助力适时调配院前急救服务资源;推进北京市院前急救供给侧改革完善,赋予院前急救服务体系更高韧性。Objective This paper analyzes the number of pre-hospital emergency ambulance trips in Beijing and predicts the number of pre-hospital emergency emergency ambulance trips in the future,hoping to provide reference for the construction and development of the pre-hospital system in Beijing.Methods Descriptive statistics were used to analyze the basic situation of the number of pre-hospital emergency ambulance trips in Beijing from 2018 to 2022.SPSS 26.0 statistical software was used to establish a time series model,and an"expert modeler"was used to automatically select the optimal model to predict the number of pre-hospital emergency ambulance trips in Beijing in 2023.Results From 2018 to 2022,the number of pre-hospital emergency ambulance trips in Beijing increased year by year,and the number of ambulance trips in 2022 was nearly twice that in 2018,the highest and lowest number of vehicles each year were in December and February;the optimal model of the time series model was automatically selected as"Winters method-additive",the model fitting degree R squared was 0.896,the stationary R squared was 0.377,and the significance P=0.642 of the statistic of Young-Box Q(18)showed good data fitting effect;the mean absolute percentage error between the predicted value and the actual value was 6.85%,indicating that the prediction ability of the model was"excellent";the number of pre-hospital emergency ambulance trips in 2023 is better predicted.Conclusion The number of pre-hospital emergency ambulance trips in Beijing has been increasing year by year,and the public health service capacity of pre-hospital emergency services has been effectively improved;the time series model fits well the variation trend of the number of pre-hospital emergency ambulance trips in Beijing and predicts it,and facilitate timely deployment of pre-hospital emergency service resources;promote the reform and improvement of the supply side of pre hospital emergency care in Beijing,and make the pre-hospital emergency care service system more resilient.

关 键 词：院前急救 时间序列模型 预测价值 

分 类 号：R459.7[医药卫生—急诊医学]