基于时序序列模型预测医院门诊人次  

Prediction of hospital outpatient visits based on time series model

作　　者：桑泉红 徐培文 Sang Quanhong;Xu Peiwen(Medical Administration Division of Shandong Provincial Third Hospital, Jinan 250031, China;The University of Adelaide, Adelaide 5000, Australia)

机构地区：[1]山东省省立三院医务部,山东济南250031 [2]澳大利亚阿德莱德大学,澳大利亚阿德莱德5000

出　　处：《中国医院统计》2022年第1期25-28,共4页Chinese Journal of Hospital Statistics

摘　　要：目的分析并预测某医院未来每年的门诊人次,为有效配比医疗资源,制定医疗工作计划和科学决策提供数据理论依据。方法采用prophet时间序列预测模型并导入统计数据,预测未来每月门诊人次,并以2019年实际数据进行比对,校验模型的准确性以及实用性。结果门诊人次季节变动以1年为1个周期,变动趋势大致相同,随季节变化而变动的特点具有明显的规律性。结论基于时间序列的prophet算法用于医院门诊量预测,方法简洁、直观,结果比较准确,可为医院领导制定医疗工作计划、科学决策提供依据。Objective To analyze and predict the number of outpatients in a hospital in the future,so as to provide data and theoretical basis for effective allocation of medical resources,formulation of medical work plan and scientific decision-making.Methods The prophet time series prediction model is adopted and the statistical data are imported to predict the number of outpatients per month in the future,and the actual data in 2019 are compared to verify the accuracy and practicability of the model.Results The seasonal variation of outpatient visits takes one year as a cycle,and the variation trend is roughly the same.The characteristics of variation with seasonal variation have obvious regularity.Conclusion The prophet algorithm based on time series is used for the prediction of hospital outpatient volume.The method is simple,intuitive and the result is relatively accurate,so it can provide the basis for hospital leaders to make medical work plans and scientific decisions.

关 键 词：门诊人次分析 医疗质量 趋势预测 

分 类 号：R195.1[医药卫生—卫生统计学]