公立医院床位利用效率及配置合理性评估研究OA北大核心CSTPCD
Evaluation of utilization efficiency and rational allocation of beds in public hospitals
目的 分析上海市某三级甲等公立医院各科室床位利用效率,为评估床位资源配置合理性提供方法学依据.方法 以上海市某三级甲等医院2023年的医疗运营数据为基础,利用床位利用模型进行可视化呈现,评价床位资源的利用效率.运用床位评价指标测算各科室床位的合理区间,得出床位调整方案.采用多层感知器神经网络模型评估床位调整方案的准确性、合理性、可行性.结果 床位利用模型显示,11个(25.00%)科室属于床位效率型,11个(25.00%)科室属于床位周转型,16个(36.36%)科室属于床位闲置型,6个(13.64%)科室属于压床型.床位评价指标显示,8个科室床位数不需改变,16个科室床位数需要适当减少,20个科室床位数需要结合实际情况增加.利用多层感知器神经网络搭建床位不变、床位减少、床位增加模型.床位不变模型的受试者工作特征曲线下面积(area under curve,AUC)=0.719,灵敏度为100.00%,特异度为40.63%.床位减少模型的AUC=0.875,灵敏度为83.33%,特异度为85.00%.床位增加模型的AUC=0.913,灵敏度为100.00%,特异度为72.22%.结论 医院整体床位利用效率较低且不同科室间床位的利用效率存在差异,通过多层感知器神经网络建立的床位增加模型评估结果与床位利用模型和床位评价指标的结果具有较好的一致性,能够为医院床位资源配置管理提供方法学依据,进而实现医院床位精细化管理.
Objective To analyze the bed utilization efficiency of various departments in a Grade-3A public hospital in Shanghai,and to provide methodological basis for evaluating the rationality of bed resources allocation.Method According to the medical operation data of a Grade-3A hospital in Shanghai in 2023,a bed utilization model was created to visually evaluate the bed efficiency.The utilization indexes were used to measure the available bed capacity in each department,which contributes to establish the bed adjustment protocol.A multi-layer perceptron neural network was employed to evaluate the protocol's specificity,rationality and feasibility.Result The bed utilization model indicated that 11(25.00%)departments shown bed efficiency,11(25.00%)departments exhibited bed weak transition,16(36.36%)departments had idle beds,and 6(13.64%)departments had occupied beds.In addition,bed efficiency analysis showed that the number of beds in 8 departments should remain the same,be reduced in 16 departments,and be expanded in 20 departments on demands.A multi-layer perceptron neural network was used to construct a system of bed-fixation model,bed-reduction model and bed-expansion model.The area under receiver operating characteristic(ROC)curve of the bed-fixation model was 0.719 with a sensitivity of 100.00%and a specificity is 40.63%.The area under ROC curve of bed-reduction model was 0.875 with a sensitivity of 83.33%and a specificity of 85.00%.The area under ROC curve of the bed-expansion model was 0.913 with a the sensitivity of 100.00%and a specificity of 72.22%.Conclusion The overall utilization efficiency of hospital beds is poor and differences occur in the utilization efficiency of beds across different departments.The evaluation outcomes of the bed-expansion model created by a multilayer perceptron neural network align well with the bed utilization model and evaluation indexes.This offers a methodological foundation for managing bed resource allocation in hospitals and achieving fine management.
袁筱祺;陈祎炜;张颜菲;孔雯;赵英英
上海市第一人民医院医务处,上海 200080上海市嘉定区疾病预防控制中心学校卫生与眼病防治科,上海 201899
预防医学
床位利用模型床位评价指标多层感知器神经网络利用效率
Bed utilization modelBed evaluation indexMulti-layer perceptron neural networkUtilization efficiency
《健康发展与政策研究》 2024 (001)
58-65 / 8
上海市申康医院发展计划项目(SHDC12022622)
评论